fix: fix channel model list is empty

* Fix ali ConvertRequest function to use baidu keyword

* Support Ali stable-diffusion-xl and wanx-v1 model

* Support Ali stable-diffusion-xl and wanx-v1 model

* Support Ali stable-diffusion-xl and wanx-v1 model

* chore: update ali constants and model ratio

---------

Co-authored-by: JustSong <songquanpeng@foxmail.com>
Co-authored-by: JustSong <39998050+songquanpeng@users.noreply.github.com>

.github/workflows/docker-image-amd64-en.yml

@@ -24,7 +24,7 @@ jobs:
         run: |
           REPO_URL=$(git config --get remote.origin.url)
           if [[ $REPO_URL == *"pro" ]]; then
-            exit 0
+            exit 1
           fi      
 
       - name: Save version info

.github/workflows/docker-image-amd64.yml

@@ -24,7 +24,7 @@ jobs:
         run: |
           REPO_URL=$(git config --get remote.origin.url)
           if [[ $REPO_URL == *"pro" ]]; then
-            exit 0
+            exit 1
           fi        
 
       - name: Save version info

.github/workflows/docker-image-arm64.yml

@@ -25,7 +25,7 @@ jobs:
         run: |
           REPO_URL=$(git config --get remote.origin.url)
           if [[ $REPO_URL == *"pro" ]]; then
-            exit 0
+            exit 1
           fi
 
       - name: Save version info

.github/workflows/linux-release.yml

@@ -24,7 +24,7 @@ jobs:
         run: |
           REPO_URL=$(git config --get remote.origin.url)
           if [[ $REPO_URL == *"pro" ]]; then
-            exit 0
+            exit 1
           fi
       - uses: actions/setup-node@v3
         with:

.github/workflows/macos-release.yml

@@ -24,7 +24,7 @@ jobs:
         run: |
           REPO_URL=$(git config --get remote.origin.url)
           if [[ $REPO_URL == *"pro" ]]; then
-            exit 0
+            exit 1
           fi
       - uses: actions/setup-node@v3
         with:

.github/workflows/windows-release.yml

@@ -27,7 +27,7 @@ jobs:
         run: |
           REPO_URL=$(git config --get remote.origin.url)
           if [[ $REPO_URL == *"pro" ]]; then
-            exit 0
+            exit 1
           fi
       - uses: actions/setup-node@v3
         with:

.gitignore

@@ -8,3 +8,4 @@ build
 logs
 data
 /web/node_modules
+cmd.md

Dockerfile

@@ -12,6 +12,10 @@ WORKDIR /web/berry
 RUN npm install
 RUN DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat VERSION) npm run build
 
+WORKDIR /web/air
+RUN npm install
+RUN DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat VERSION) npm run build
+
 FROM golang AS builder2
 
 ENV GO111MODULE=on \

README.en.md

@@ -241,17 +241,19 @@ If the channel ID is not provided, load balancing will be used to distribute the
     + Example: `SESSION_SECRET=random_string`
 3. `SQL_DSN`: When set, the specified database will be used instead of SQLite. Please use MySQL version 8.0.
     + Example: `SQL_DSN=root:123456@tcp(localhost:3306)/oneapi`
-4. `FRONTEND_BASE_URL`: When set, the specified frontend address will be used instead of the backend address.
+4. `LOG_SQL_DSN`: When set, a separate database will be used for the `logs` table; please use MySQL or PostgreSQL.
+    + Example: `LOG_SQL_DSN=root:123456@tcp(localhost:3306)/oneapi-logs`
+5. `FRONTEND_BASE_URL`: When set, the specified frontend address will be used instead of the backend address.
     + Example: `FRONTEND_BASE_URL=https://openai.justsong.cn`
-5. `SYNC_FREQUENCY`: When set, the system will periodically sync configurations from the database, with the unit in seconds. If not set, no sync will happen.
+6. `SYNC_FREQUENCY`: When set, the system will periodically sync configurations from the database, with the unit in seconds. If not set, no sync will happen.
     + Example: `SYNC_FREQUENCY=60`
-6. `NODE_TYPE`: When set, specifies the node type. Valid values are `master` and `slave`. If not set, it defaults to `master`.
+7. `NODE_TYPE`: When set, specifies the node type. Valid values are `master` and `slave`. If not set, it defaults to `master`.
     + Example: `NODE_TYPE=slave`
-7. `CHANNEL_UPDATE_FREQUENCY`: When set, it periodically updates the channel balances, with the unit in minutes. If not set, no update will happen.
+8. `CHANNEL_UPDATE_FREQUENCY`: When set, it periodically updates the channel balances, with the unit in minutes. If not set, no update will happen.
     + Example: `CHANNEL_UPDATE_FREQUENCY=1440`
-8. `CHANNEL_TEST_FREQUENCY`: When set, it periodically tests the channels, with the unit in minutes. If not set, no test will happen.
+9. `CHANNEL_TEST_FREQUENCY`: When set, it periodically tests the channels, with the unit in minutes. If not set, no test will happen.
     + Example: `CHANNEL_TEST_FREQUENCY=1440`
-9. `POLLING_INTERVAL`: The time interval (in seconds) between requests when updating channel balances and testing channel availability. Default is no interval.
+10. `POLLING_INTERVAL`: The time interval (in seconds) between requests when updating channel balances and testing channel availability. Default is no interval.
     + Example: `POLLING_INTERVAL=5`
 
 ### Command Line Parameters

README.ja.md

@@ -242,17 +242,18 @@ graph LR
     + 例: `SESSION_SECRET=random_string`
 3. `SQL_DSN`: 設定すると、SQLite の代わりに指定したデータベースが使用されます。MySQL バージョン 8.0 を使用してください。
     + 例: `SQL_DSN=root:123456@tcp(localhost:3306)/oneapi`
-4. `FRONTEND_BASE_URL`: 設定されると、バックエンドアドレスではなく、指定されたフロントエンドアドレスが使われる。
+4. `LOG_SQL_DSN`: を設定すると、`logs`テーブルには独立したデータベースが使用されます。MySQLまたはPostgreSQLを使用してください。
+5. `FRONTEND_BASE_URL`: 設定されると、バックエンドアドレスではなく、指定されたフロントエンドアドレスが使われる。
     + 例: `FRONTEND_BASE_URL=https://openai.justsong.cn`
-5. `SYNC_FREQUENCY`: 設定された場合、システムは定期的にデータベースからコンフィグを秒単位で同期する。設定されていない場合、同期は行われません。
+6. `SYNC_FREQUENCY`: 設定された場合、システムは定期的にデータベースからコンフィグを秒単位で同期する。設定されていない場合、同期は行われません。
     + 例: `SYNC_FREQUENCY=60`
-6. `NODE_TYPE`: 設定すると、ノードのタイプを指定する。有効な値は `master` と `slave` である。設定されていない場合、デフォルトは `master`。
+7. `NODE_TYPE`: 設定すると、ノードのタイプを指定する。有効な値は `master` と `slave` である。設定されていない場合、デフォルトは `master`。
     + 例: `NODE_TYPE=slave`
-7. `CHANNEL_UPDATE_FREQUENCY`: 設定すると、チャンネル残高を分単位で定期的に更新する。設定されていない場合、更新は行われません。
+8. `CHANNEL_UPDATE_FREQUENCY`: 設定すると、チャンネル残高を分単位で定期的に更新する。設定されていない場合、更新は行われません。
     + 例: `CHANNEL_UPDATE_FREQUENCY=1440`
-8. `CHANNEL_TEST_FREQUENCY`: 設定すると、チャンネルを定期的にテストする。設定されていない場合、テストは行われません。
+9. `CHANNEL_TEST_FREQUENCY`: 設定すると、チャンネルを定期的にテストする。設定されていない場合、テストは行われません。
     + 例: `CHANNEL_TEST_FREQUENCY=1440`
-9. `POLLING_INTERVAL`: チャネル残高の更新とチャネルの可用性をテストするときのリクエスト間の時間間隔 (秒)。デフォルトは間隔なし。
+10. `POLLING_INTERVAL`: チャネル残高の更新とチャネルの可用性をテストするときのリクエスト間の時間間隔 (秒)。デフォルトは間隔なし。
     + 例: `POLLING_INTERVAL=5`
 
 ### コマンドラインパラメータ

README.md

@@ -81,13 +81,14 @@ _✨ 通过标准的 OpenAI API 格式访问所有的大模型，开箱即用
    + [x] [Groq](https://wow.groq.com/)
    + [x] [Ollama](https://github.com/ollama/ollama)
    + [x] [零一万物](https://platform.lingyiwanwu.com/)
+   + [x] [阶跃星辰](https://platform.stepfun.com/)
 2. 支持配置镜像以及众多[第三方代理服务](https://iamazing.cn/page/openai-api-third-party-services)。
 3. 支持通过**负载均衡**的方式访问多个渠道。
 4. 支持 **stream 模式**，可以通过流式传输实现打字机效果。
 5. 支持**多机部署**，[详见此处](#多机部署)。
-6. 支持**令牌管理**，设置令牌的过期时间和额度。
+6. 支持**令牌管理**，设置令牌的过期时间、额度、允许的 IP 范围以及允许的模型访问。
 7. 支持**兑换码管理**，支持批量生成和导出兑换码，可使用兑换码为账户进行充值。
-8. 支持**通道管理**，批量创建通道。
+8. 支持**渠道管理**，批量创建渠道。
 9. 支持**用户分组**以及**渠道分组**，支持为不同分组设置不同的倍率。
 10. 支持渠道**设置模型列表**。
 11. 支持**查看额度明细**。
@@ -101,10 +102,11 @@ _✨ 通过标准的 OpenAI API 格式访问所有的大模型，开箱即用
 19. 支持丰富的**自定义**设置，
     1. 支持自定义系统名称，logo 以及页脚。
     2. 支持自定义首页和关于页面，可以选择使用 HTML & Markdown 代码进行自定义，或者使用一个单独的网页通过 iframe 嵌入。
-20. 支持通过系统访问令牌访问管理 API（bearer token，用以替代 cookie，你可以自行抓包来查看 API 的用法）。
+20. 支持通过系统访问令牌调用管理 API，进而**在无需二开的情况下扩展和自定义** One API 的功能，详情请参考此处 [API 文档](./docs/API.md)。。
 21. 支持 Cloudflare Turnstile 用户校验。
 22. 支持用户管理，支持**多种用户登录注册方式**：
     + 邮箱登录注册（支持注册邮箱白名单）以及通过邮箱进行密码重置。
+    + 支持使用飞书进行授权登录。
     + [GitHub 开放授权](https://github.com/settings/applications/new)。
     + 微信公众号授权（需要额外部署 [WeChat Server](https://github.com/songquanpeng/wechat-server)）。
 23. 支持主题切换，设置环境变量 `THEME` 即可，默认为 `default`，欢迎 PR 更多主题，具体参考[此处](./web/README.md)。
@@ -349,38 +351,40 @@ graph LR
      + `SQL_MAX_OPEN_CONNS`：最大打开连接数，默认为 `1000`。
        + 如果报错 `Error 1040: Too many connections`，请适当减小该值。
      + `SQL_CONN_MAX_LIFETIME`：连接的最大生命周期，默认为 `60`，单位分钟。
-4. `FRONTEND_BASE_URL`：设置之后将重定向页面请求到指定的地址，仅限从服务器设置。
+4. `LOG_SQL_DSN`：设置之后将为 `logs` 表使用独立的数据库，请使用 MySQL 或 PostgreSQL。
+5. `FRONTEND_BASE_URL`：设置之后将重定向页面请求到指定的地址，仅限从服务器设置。
    + 例子：`FRONTEND_BASE_URL=https://openai.justsong.cn`
-5. `MEMORY_CACHE_ENABLED`：启用内存缓存，会导致用户额度的更新存在一定的延迟，可选值为 `true` 和 `false`，未设置则默认为 `false`。
+6. `MEMORY_CACHE_ENABLED`：启用内存缓存，会导致用户额度的更新存在一定的延迟，可选值为 `true` 和 `false`，未设置则默认为 `false`。
    + 例子：`MEMORY_CACHE_ENABLED=true`
-6. `SYNC_FREQUENCY`：在启用缓存的情况下与数据库同步配置的频率，单位为秒，默认为 `600` 秒。
+7. `SYNC_FREQUENCY`：在启用缓存的情况下与数据库同步配置的频率，单位为秒，默认为 `600` 秒。
    + 例子：`SYNC_FREQUENCY=60`
-7. `NODE_TYPE`：设置之后将指定节点类型，可选值为 `master` 和 `slave`，未设置则默认为 `master`。
+8. `NODE_TYPE`：设置之后将指定节点类型，可选值为 `master` 和 `slave`，未设置则默认为 `master`。
    + 例子：`NODE_TYPE=slave`
-8. `CHANNEL_UPDATE_FREQUENCY`：设置之后将定期更新渠道余额，单位为分钟，未设置则不进行更新。
+9. `CHANNEL_UPDATE_FREQUENCY`：设置之后将定期更新渠道余额，单位为分钟，未设置则不进行更新。
    + 例子：`CHANNEL_UPDATE_FREQUENCY=1440`
-9. `CHANNEL_TEST_FREQUENCY`：设置之后将定期检查渠道，单位为分钟，未设置则不进行检查。
+10. `CHANNEL_TEST_FREQUENCY`：设置之后将定期检查渠道，单位为分钟，未设置则不进行检查。
    + 例子：`CHANNEL_TEST_FREQUENCY=1440`
-10. `POLLING_INTERVAL`：批量更新渠道余额以及测试可用性时的请求间隔，单位为秒，默认无间隔。
+11. `POLLING_INTERVAL`：批量更新渠道余额以及测试可用性时的请求间隔，单位为秒，默认无间隔。
     + 例子：`POLLING_INTERVAL=5`
-11. `BATCH_UPDATE_ENABLED`：启用数据库批量更新聚合，会导致用户额度的更新存在一定的延迟可选值为 `true` 和 `false`，未设置则默认为 `false`。
+12. `BATCH_UPDATE_ENABLED`：启用数据库批量更新聚合，会导致用户额度的更新存在一定的延迟可选值为 `true` 和 `false`，未设置则默认为 `false`。
     + 例子：`BATCH_UPDATE_ENABLED=true`
     + 如果你遇到了数据库连接数过多的问题，可以尝试启用该选项。
-12. `BATCH_UPDATE_INTERVAL=5`：批量更新聚合的时间间隔，单位为秒，默认为 `5`。
+13. `BATCH_UPDATE_INTERVAL=5`：批量更新聚合的时间间隔，单位为秒，默认为 `5`。
     + 例子：`BATCH_UPDATE_INTERVAL=5`
-13. 请求频率限制：
+14. 请求频率限制：
     + `GLOBAL_API_RATE_LIMIT`：全局 API 速率限制（除中继请求外），单 ip 三分钟内的最大请求数，默认为 `180`。
     + `GLOBAL_WEB_RATE_LIMIT`：全局 Web 速率限制，单 ip 三分钟内的最大请求数，默认为 `60`。
-14. 编码器缓存设置：
+15. 编码器缓存设置：
     + `TIKTOKEN_CACHE_DIR`：默认程序启动时会联网下载一些通用的词元的编码，如：`gpt-3.5-turbo`，在一些网络环境不稳定，或者离线情况，可能会导致启动有问题，可以配置此目录缓存数据，可迁移到离线环境。
     + `DATA_GYM_CACHE_DIR`：目前该配置作用与 `TIKTOKEN_CACHE_DIR` 一致，但是优先级没有它高。
-15. `RELAY_TIMEOUT`：中继超时设置，单位为秒，默认不设置超时时间。
-16. `SQLITE_BUSY_TIMEOUT`：SQLite 锁等待超时设置，单位为毫秒，默认 `3000`。
-17. `GEMINI_SAFETY_SETTING`：Gemini 的安全设置，默认 `BLOCK_NONE`。
-18. `THEME`：系统的主题设置，默认为 `default`，具体可选值参考[此处](./web/README.md)。
-19. `ENABLE_METRIC`：是否根据请求成功率禁用渠道，默认不开启，可选值为 `true` 和 `false`。
-20. `METRIC_QUEUE_SIZE`：请求成功率统计队列大小，默认为 `10`。
-21. `METRIC_SUCCESS_RATE_THRESHOLD`：请求成功率阈值，默认为 `0.8`。
+16. `RELAY_TIMEOUT`：中继超时设置，单位为秒，默认不设置超时时间。
+17. `SQLITE_BUSY_TIMEOUT`：SQLite 锁等待超时设置，单位为毫秒，默认 `3000`。
+18. `GEMINI_SAFETY_SETTING`：Gemini 的安全设置，默认 `BLOCK_NONE`。
+19. `THEME`：系统的主题设置，默认为 `default`，具体可选值参考[此处](./web/README.md)。
+20. `ENABLE_METRIC`：是否根据请求成功率禁用渠道，默认不开启，可选值为 `true` 和 `false`。
+21. `METRIC_QUEUE_SIZE`：请求成功率统计队列大小，默认为 `10`。
+22. `METRIC_SUCCESS_RATE_THRESHOLD`：请求成功率阈值，默认为 `0.8`。
+23. `INITIAL_ROOT_TOKEN`：如果设置了该值，则在系统首次启动时会自动创建一个值为该环境变量值的 root 用户令牌。
 
 ### 命令行参数
 1. `--port <port_number>`: 指定服务器监听的端口号，默认为 `3000`。
@@ -419,7 +423,7 @@ https://openai.justsong.cn
    + 检查你的接口地址和 API Key 有没有填对。
    + 检查是否启用了 HTTPS，浏览器会拦截 HTTPS 域名下的 HTTP 请求。
 6. 报错：`当前分组负载已饱和，请稍后再试`
-   + 上游通道 429 了。
+   + 上游渠道 429 了。
 7. 升级之后我的数据会丢失吗？
    + 如果使用 MySQL，不会。
    + 如果使用 SQLite，需要按照我所给的部署命令挂载 volume 持久化 one-api.db 数据库文件，否则容器重启后数据会丢失。
@@ -427,8 +431,8 @@ https://openai.justsong.cn
    + 一般情况下不需要，系统将在初始化的时候自动调整。
    + 如果需要的话，我会在更新日志中说明，并给出脚本。
 9. 手动修改数据库后报错：`数据库一致性已被破坏，请联系管理员`？
-   + 这是检测到 ability 表里有些记录的通道 id 是不存在的，这大概率是因为你删了 channel 表里的记录但是没有同步在 ability 表里清理无效的通道。
-   + 对于每一个通道，其所支持的模型都需要有一个专门的 ability 表的记录，表示该通道支持该模型。
+   + 这是检测到 ability 表里有些记录的渠道 id 是不存在的，这大概率是因为你删了 channel 表里的记录但是没有同步在 ability 表里清理无效的渠道。
+   + 对于每一个渠道，其所支持的模型都需要有一个专门的 ability 表的记录，表示该渠道支持该模型。
 
 ## 相关项目
 * [FastGPT](https://github.com/labring/FastGPT): 基于 LLM 大语言模型的知识库问答系统

common/config/config.go

@@ -66,6 +66,9 @@ var SMTPToken = ""
 var GitHubClientId = ""
 var GitHubClientSecret = ""
 
+var LarkClientId = ""
+var LarkClientSecret = ""
+
 var WeChatServerAddress = ""
 var WeChatServerToken = ""
 var WeChatAccountQRCodeImageURL = ""
@@ -107,6 +110,7 @@ var Theme = env.String("THEME", "default")
 var ValidThemes = map[string]bool{
 	"default": true,
 	"berry":   true,
+	"air":     true,
 }
 
 // All duration's unit is seconds
@@ -135,3 +139,5 @@ var MetricQueueSize = env.Int("METRIC_QUEUE_SIZE", 10)
 var MetricSuccessRateThreshold = env.Float64("METRIC_SUCCESS_RATE_THRESHOLD", 0.8)
 var MetricSuccessChanSize = env.Int("METRIC_SUCCESS_CHAN_SIZE", 1024)
 var MetricFailChanSize = env.Int("METRIC_FAIL_CHAN_SIZE", 128)
+
+var InitialRootToken = os.Getenv("INITIAL_ROOT_TOKEN")

common/constants.go

@@ -71,6 +71,7 @@ const (
 	ChannelTypeGroq
 	ChannelTypeOllama
 	ChannelTypeLingYiWanWu
+	ChannelTypeStepFun
 
 	ChannelTypeDummy
 )
@@ -108,6 +109,7 @@ var ChannelBaseURLs = []string{
 	"https://api.groq.com/openai",               // 29
 	"http://localhost:11434",                    // 30
 	"https://api.lingyiwanwu.com",               // 31
+	"https://api.stepfun.com",                   // 32
 }
 
 const (

common/conv/any.go

@@ -0,0 +1,6 @@
+package conv
+
+func AsString(v any) string {
+	str, _ := v.(string)
+	return str
+}

common/model-ratio.go

@@ -4,7 +4,6 @@ import (
 	"encoding/json"
 	"github.com/songquanpeng/one-api/common/logger"
 	"strings"
-	"time"
 )
 
 const (
@@ -31,7 +30,7 @@ var ModelRatio = map[string]float64{
 	"gpt-4-0125-preview":      5,    // $0.01 / 1K tokens
 	"gpt-4-turbo-preview":     5,    // $0.01 / 1K tokens
 	"gpt-4-vision-preview":    5,    // $0.01 / 1K tokens
-	"gpt-3.5-turbo":           0.75, // $0.0015 / 1K tokens
+	"gpt-3.5-turbo":           0.25, // $0.0005 / 1K tokens
 	"gpt-3.5-turbo-0301":      0.75,
 	"gpt-3.5-turbo-0613":      0.75,
 	"gpt-3.5-turbo-16k":       1.5, // $0.003 / 1K tokens
@@ -73,30 +72,47 @@ var ModelRatio = map[string]float64{
 	"claude-3-sonnet-20240229": 3.0 / 1000 * USD,
 	"claude-3-opus-20240229":   15.0 / 1000 * USD,
 	// https://cloud.baidu.com/doc/WENXINWORKSHOP/s/hlrk4akp7
-	"ERNIE-Bot":         0.8572,     // ￥0.012 / 1k tokens
-	"ERNIE-Bot-turbo":   0.5715,     // ￥0.008 / 1k tokens
-	"ERNIE-Bot-4":       0.12 * RMB, // ￥0.12 / 1k tokens
-	"ERNIE-Bot-8k":      0.024 * RMB,
-	"Embedding-V1":      0.1429, // ￥0.002 / 1k tokens
-	"bge-large-zh":      0.002 * RMB,
-	"bge-large-en":      0.002 * RMB,
-	"bge-large-8k":      0.002 * RMB,
-	"PaLM-2":            1,
-	"gemini-pro":        1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
-	"gemini-pro-vision": 1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
+	"ERNIE-4.0-8K":       0.120 * RMB,
+	"ERNIE-3.5-8K":       0.012 * RMB,
+	"ERNIE-3.5-8K-0205":  0.024 * RMB,
+	"ERNIE-3.5-8K-1222":  0.012 * RMB,
+	"ERNIE-Bot-8K":       0.024 * RMB,
+	"ERNIE-3.5-4K-0205":  0.012 * RMB,
+	"ERNIE-Speed-8K":     0.004 * RMB,
+	"ERNIE-Speed-128K":   0.004 * RMB,
+	"ERNIE-Lite-8K-0922": 0.008 * RMB,
+	"ERNIE-Lite-8K-0308": 0.003 * RMB,
+	"ERNIE-Tiny-8K":      0.001 * RMB,
+	"BLOOMZ-7B":          0.004 * RMB,
+	"Embedding-V1":       0.002 * RMB,
+	"bge-large-zh":       0.002 * RMB,
+	"bge-large-en":       0.002 * RMB,
+	"tao-8k":             0.002 * RMB,
+	// https://ai.google.dev/pricing
+	"PaLM-2":                    1,
+	"gemini-pro":                1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
+	"gemini-pro-vision":         1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
+	"gemini-1.0-pro-vision-001": 1,
+	"gemini-1.0-pro-001":        1,
+	"gemini-1.5-pro":            1,
 	// https://open.bigmodel.cn/pricing
 	"glm-4":                     0.1 * RMB,
 	"glm-4v":                    0.1 * RMB,
 	"glm-3-turbo":               0.005 * RMB,
+	"embedding-2":               0.0005 * RMB,
 	"chatglm_turbo":             0.3572, // ￥0.005 / 1k tokens
 	"chatglm_pro":               0.7143, // ￥0.01 / 1k tokens
 	"chatglm_std":               0.3572, // ￥0.005 / 1k tokens
 	"chatglm_lite":              0.1429, // ￥0.002 / 1k tokens
-	"qwen-turbo":                0.5715, // ￥0.008 / 1k tokens  // https://help.aliyun.com/zh/dashscope/developer-reference/tongyi-thousand-questions-metering-and-billing
+	// https://help.aliyun.com/zh/dashscope/developer-reference/tongyi-thousand-questions-metering-and-billing
+	"qwen-turbo":                0.5715, // ￥0.008 / 1k tokens
 	"qwen-plus":                 1.4286, // ￥0.02 / 1k tokens
 	"qwen-max":                  1.4286, // ￥0.02 / 1k tokens
 	"qwen-max-longcontext":      1.4286, // ￥0.02 / 1k tokens
 	"text-embedding-v1":         0.05,   // ￥0.0007 / 1k tokens
+	"ali-stable-diffusion-xl":   8,
+	"ali-stable-diffusion-v1.5": 8,
+	"wanx-v1":                   8,
 	"SparkDesk":                 1.2858, // ￥0.018 / 1k tokens
 	"SparkDesk-v1.1":            1.2858, // ￥0.018 / 1k tokens
 	"SparkDesk-v2.1":            1.2858, // ￥0.018 / 1k tokens
@@ -134,9 +150,9 @@ var ModelRatio = map[string]float64{
 	"mixtral-8x7b-32768": 0.27 / 1000 * USD,
 	"gemma-7b-it":        0.1 / 1000 * USD,
 	// https://platform.lingyiwanwu.com/docs#-计费单元
-	"yi-34b-chat-0205": 2.5 / 1000000 * RMB,
-	"yi-34b-chat-200k": 12.0 / 1000000 * RMB,
-	"yi-vl-plus":       6.0 / 1000000 * RMB,
+	"yi-34b-chat-0205": 2.5 / 1000 * RMB,
+	"yi-34b-chat-200k": 12.0 / 1000 * RMB,
+	"yi-vl-plus":       6.0 / 1000 * RMB,
 }
 
 var CompletionRatio = map[string]float64{}
@@ -224,7 +240,7 @@ func GetCompletionRatio(name string) float64 {
 		return ratio
 	}
 	if strings.HasPrefix(name, "gpt-3.5") {
-		if strings.HasSuffix(name, "0125") {
+		if name == "gpt-3.5-turbo" || strings.HasSuffix(name, "0125") {
 			// https://openai.com/blog/new-embedding-models-and-api-updates
 			// Updated GPT-3.5 Turbo model and lower pricing
 			return 3
@@ -232,15 +248,6 @@ func GetCompletionRatio(name string) float64 {
 		if strings.HasSuffix(name, "1106") {
 			return 2
 		}
-		if name == "gpt-3.5-turbo" || name == "gpt-3.5-turbo-16k" {
-			// TODO: clear this after 2023-12-11
-			now := time.Now()
-			// https://platform.openai.com/docs/models/continuous-model-upgrades
-			// if after 2023-12-11, use 2
-			if now.After(time.Date(2023, 12, 11, 0, 0, 0, 0, time.UTC)) {
-				return 2
-			}
-		}
 		return 4.0 / 3.0
 	}
 	if strings.HasPrefix(name, "gpt-4") {
@@ -258,6 +265,9 @@ func GetCompletionRatio(name string) float64 {
 	if strings.HasPrefix(name, "mistral-") {
 		return 3
 	}
+	if strings.HasPrefix(name, "gemini-") {
+		return 3
+	}
 	switch name {
 	case "llama2-70b-4096":
 		return 0.8 / 0.7

common/network/ip.go

@@ -0,0 +1,52 @@
+package network
+
+import (
+	"context"
+	"fmt"
+	"github.com/songquanpeng/one-api/common/logger"
+	"net"
+	"strings"
+)
+
+func splitSubnets(subnets string) []string {
+	res := strings.Split(subnets, ",")
+	for i := 0; i < len(res); i++ {
+		res[i] = strings.TrimSpace(res[i])
+	}
+	return res
+}
+
+func isValidSubnet(subnet string) error {
+	_, _, err := net.ParseCIDR(subnet)
+	if err != nil {
+		return fmt.Errorf("failed to parse subnet: %w", err)
+	}
+	return nil
+}
+
+func isIpInSubnet(ctx context.Context, ip string, subnet string) bool {
+	_, ipNet, err := net.ParseCIDR(subnet)
+	if err != nil {
+		logger.Errorf(ctx, "failed to parse subnet: %s", err.Error())
+		return false
+	}
+	return ipNet.Contains(net.ParseIP(ip))
+}
+
+func IsValidSubnets(subnets string) error {
+	for _, subnet := range splitSubnets(subnets) {
+		if err := isValidSubnet(subnet); err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+func IsIpInSubnets(ctx context.Context, ip string, subnets string) bool {
+	for _, subnet := range splitSubnets(subnets) {
+		if isIpInSubnet(ctx, ip, subnet) {
+			return true
+		}
+	}
+	return false
+}

common/network/ip_test.go

@@ -0,0 +1,19 @@
+package network
+
+import (
+	"context"
+	"testing"
+
+	. "github.com/smartystreets/goconvey/convey"
+)
+
+func TestIsIpInSubnet(t *testing.T) {
+	ctx := context.Background()
+	ip1 := "192.168.0.5"
+	ip2 := "125.216.250.89"
+	subnet := "192.168.0.0/24"
+	Convey("TestIsIpInSubnet", t, func() {
+		So(isIpInSubnet(ctx, ip1, subnet), ShouldBeTrue)
+		So(isIpInSubnet(ctx, ip2, subnet), ShouldBeFalse)
+	})
+}

controller/auth/github.go

@@ -1,4 +1,4 @@
-package controller
+package auth
 
 import (
 	"bytes"
@@ -11,6 +11,7 @@ import (
 	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/logger"
+	"github.com/songquanpeng/one-api/controller"
 	"github.com/songquanpeng/one-api/model"
 	"net/http"
 	"strconv"
@@ -159,7 +160,7 @@ func GitHubOAuth(c *gin.Context) {
 		})
 		return
 	}
-	setupLogin(&user, c)
+	controller.SetupLogin(&user, c)
 }
 
 func GitHubBind(c *gin.Context) {

controller/auth/lark.go

@@ -0,0 +1,201 @@
+package auth
+
+import (
+	"bytes"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"github.com/gin-contrib/sessions"
+	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/common"
+	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/common/logger"
+	"github.com/songquanpeng/one-api/controller"
+	"github.com/songquanpeng/one-api/model"
+	"net/http"
+	"strconv"
+	"time"
+)
+
+type LarkOAuthResponse struct {
+	AccessToken string `json:"access_token"`
+}
+
+type LarkUser struct {
+	Name   string `json:"name"`
+	OpenID string `json:"open_id"`
+}
+
+func getLarkUserInfoByCode(code string) (*LarkUser, error) {
+	if code == "" {
+		return nil, errors.New("无效的参数")
+	}
+	values := map[string]string{
+		"client_id":     config.LarkClientId,
+		"client_secret": config.LarkClientSecret,
+		"code":          code,
+		"grant_type":    "authorization_code",
+		"redirect_uri":  fmt.Sprintf("%s/oauth/lark", config.ServerAddress),
+	}
+	jsonData, err := json.Marshal(values)
+	if err != nil {
+		return nil, err
+	}
+	req, err := http.NewRequest("POST", "https://passport.feishu.cn/suite/passport/oauth/token", bytes.NewBuffer(jsonData))
+	if err != nil {
+		return nil, err
+	}
+	req.Header.Set("Content-Type", "application/json")
+	req.Header.Set("Accept", "application/json")
+	client := http.Client{
+		Timeout: 5 * time.Second,
+	}
+	res, err := client.Do(req)
+	if err != nil {
+		logger.SysLog(err.Error())
+		return nil, errors.New("无法连接至飞书服务器，请稍后重试！")
+	}
+	defer res.Body.Close()
+	var oAuthResponse LarkOAuthResponse
+	err = json.NewDecoder(res.Body).Decode(&oAuthResponse)
+	if err != nil {
+		return nil, err
+	}
+	req, err = http.NewRequest("GET", "https://passport.feishu.cn/suite/passport/oauth/userinfo", nil)
+	if err != nil {
+		return nil, err
+	}
+	req.Header.Set("Authorization", fmt.Sprintf("Bearer %s", oAuthResponse.AccessToken))
+	res2, err := client.Do(req)
+	if err != nil {
+		logger.SysLog(err.Error())
+		return nil, errors.New("无法连接至飞书服务器，请稍后重试！")
+	}
+	var larkUser LarkUser
+	err = json.NewDecoder(res2.Body).Decode(&larkUser)
+	if err != nil {
+		return nil, err
+	}
+	return &larkUser, nil
+}
+
+func LarkOAuth(c *gin.Context) {
+	session := sessions.Default(c)
+	state := c.Query("state")
+	if state == "" || session.Get("oauth_state") == nil || state != session.Get("oauth_state").(string) {
+		c.JSON(http.StatusForbidden, gin.H{
+			"success": false,
+			"message": "state is empty or not same",
+		})
+		return
+	}
+	username := session.Get("username")
+	if username != nil {
+		LarkBind(c)
+		return
+	}
+	code := c.Query("code")
+	larkUser, err := getLarkUserInfoByCode(code)
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": err.Error(),
+		})
+		return
+	}
+	user := model.User{
+		LarkId: larkUser.OpenID,
+	}
+	if model.IsLarkIdAlreadyTaken(user.LarkId) {
+		err := user.FillUserByLarkId()
+		if err != nil {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": err.Error(),
+			})
+			return
+		}
+	} else {
+		if config.RegisterEnabled {
+			user.Username = "lark_" + strconv.Itoa(model.GetMaxUserId()+1)
+			if larkUser.Name != "" {
+				user.DisplayName = larkUser.Name
+			} else {
+				user.DisplayName = "Lark User"
+			}
+			user.Role = common.RoleCommonUser
+			user.Status = common.UserStatusEnabled
+
+			if err := user.Insert(0); err != nil {
+				c.JSON(http.StatusOK, gin.H{
+					"success": false,
+					"message": err.Error(),
+				})
+				return
+			}
+		} else {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "管理员关闭了新用户注册",
+			})
+			return
+		}
+	}
+
+	if user.Status != common.UserStatusEnabled {
+		c.JSON(http.StatusOK, gin.H{
+			"message": "用户已被封禁",
+			"success": false,
+		})
+		return
+	}
+	controller.SetupLogin(&user, c)
+}
+
+func LarkBind(c *gin.Context) {
+	code := c.Query("code")
+	larkUser, err := getLarkUserInfoByCode(code)
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": err.Error(),
+		})
+		return
+	}
+	user := model.User{
+		LarkId: larkUser.OpenID,
+	}
+	if model.IsLarkIdAlreadyTaken(user.LarkId) {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": "该飞书账户已被绑定",
+		})
+		return
+	}
+	session := sessions.Default(c)
+	id := session.Get("id")
+	// id := c.GetInt("id")  // critical bug!
+	user.Id = id.(int)
+	err = user.FillUserById()
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": err.Error(),
+		})
+		return
+	}
+	user.LarkId = larkUser.OpenID
+	err = user.Update(false)
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": err.Error(),
+		})
+		return
+	}
+	c.JSON(http.StatusOK, gin.H{
+		"success": true,
+		"message": "bind",
+	})
+	return
+}

controller/auth/wechat.go

@@ -1,4 +1,4 @@
-package controller
+package auth
 
 import (
 	"encoding/json"
@@ -7,6 +7,7 @@ import (
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/controller"
 	"github.com/songquanpeng/one-api/model"
 	"net/http"
 	"strconv"
@@ -109,7 +110,7 @@ func WeChatAuth(c *gin.Context) {
 		})
 		return
 	}
-	setupLogin(&user, c)
+	controller.SetupLogin(&user, c)
 }
 
 func WeChatBind(c *gin.Context) {

controller/channel-test.go

@@ -197,7 +197,7 @@ func testChannels(notify bool, scope string) error {
 		testAllChannelsRunning = false
 		testAllChannelsLock.Unlock()
 		if notify {
-			err := message.Notify(message.ByAll, "通道测试完成", "", "通道测试完成，如果没有收到禁用通知，说明所有通道都正常")
+			err := message.Notify(message.ByAll, "渠道测试完成", "", "渠道测试完成，如果没有收到禁用通知，说明所有渠道都正常")
 			if err != nil {
 				logger.SysError(fmt.Sprintf("failed to send email: %s", err.Error()))
 			}

controller/misc.go

@@ -23,6 +23,7 @@ func GetStatus(c *gin.Context) {
 			"email_verification":  config.EmailVerificationEnabled,
 			"github_oauth":        config.GitHubOAuthEnabled,
 			"github_client_id":    config.GitHubClientId,
+			"lark_client_id":      config.LarkClientId,
 			"system_name":         config.SystemName,
 			"logo":                config.Logo,
 			"footer_html":         config.Footer,

controller/model.go

@@ -4,12 +4,14 @@ import (
 	"fmt"
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
+	"github.com/songquanpeng/one-api/model"
 	"github.com/songquanpeng/one-api/relay/channel/openai"
 	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/helper"
 	relaymodel "github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/util"
 	"net/http"
+	"strings"
 )
 
 // https://platform.openai.com/docs/api-reference/models/list
@@ -39,8 +41,8 @@ type OpenAIModels struct {
 	Parent     *string                 `json:"parent"`
 }
 
-var openAIModels []OpenAIModels
-var openAIModelsMap map[string]OpenAIModels
+var models []OpenAIModels
+var modelsMap map[string]OpenAIModels
 var channelId2Models map[int][]string
 
 func init() {
@@ -68,7 +70,7 @@ func init() {
 		channelName := adaptor.GetChannelName()
 		modelNames := adaptor.GetModelList()
 		for _, modelName := range modelNames {
-			openAIModels = append(openAIModels, OpenAIModels{
+			models = append(models, OpenAIModels{
 				Id:         modelName,
 				Object:     "model",
 				Created:    1626777600,
@@ -85,7 +87,7 @@ func init() {
 		}
 		channelName, channelModelList := openai.GetCompatibleChannelMeta(channelType)
 		for _, modelName := range channelModelList {
-			openAIModels = append(openAIModels, OpenAIModels{
+			models = append(models, OpenAIModels{
 				Id:         modelName,
 				Object:     "model",
 				Created:    1626777600,
@@ -96,9 +98,9 @@ func init() {
 			})
 		}
 	}
-	openAIModelsMap = make(map[string]OpenAIModels)
-	for _, model := range openAIModels {
-		openAIModelsMap[model.Id] = model
+	modelsMap = make(map[string]OpenAIModels)
+	for _, model := range models {
+		modelsMap[model.Id] = model
 	}
 	channelId2Models = make(map[int][]string)
 	for i := 1; i < common.ChannelTypeDummy; i++ {
@@ -119,16 +121,55 @@ func DashboardListModels(c *gin.Context) {
 	})
 }
 
-func ListModels(c *gin.Context) {
+func ListAllModels(c *gin.Context) {
 	c.JSON(200, gin.H{
 		"object": "list",
-		"data":   openAIModels,
+		"data":   models,
+	})
+}
+
+func ListModels(c *gin.Context) {
+	ctx := c.Request.Context()
+	var availableModels []string
+	if c.GetString("available_models") != "" {
+		availableModels = strings.Split(c.GetString("available_models"), ",")
+	} else {
+		userId := c.GetInt("id")
+		userGroup, _ := model.CacheGetUserGroup(userId)
+		availableModels, _ = model.CacheGetGroupModels(ctx, userGroup)
+	}
+	modelSet := make(map[string]bool)
+	for _, availableModel := range availableModels {
+		modelSet[availableModel] = true
+	}
+	availableOpenAIModels := make([]OpenAIModels, 0)
+	for _, model := range models {
+		if _, ok := modelSet[model.Id]; ok {
+			modelSet[model.Id] = false
+			availableOpenAIModels = append(availableOpenAIModels, model)
+		}
+	}
+	for modelName, ok := range modelSet {
+		if ok {
+			availableOpenAIModels = append(availableOpenAIModels, OpenAIModels{
+				Id:      modelName,
+				Object:  "model",
+				Created: 1626777600,
+				OwnedBy: "custom",
+				Root:    modelName,
+				Parent:  nil,
+			})
+		}
+	}
+	c.JSON(200, gin.H{
+		"object": "list",
+		"data":   availableOpenAIModels,
 	})
 }
 
 func RetrieveModel(c *gin.Context) {
 	modelId := c.Param("model")
-	if model, ok := openAIModelsMap[modelId]; ok {
+	if model, ok := modelsMap[modelId]; ok {
 		c.JSON(200, model)
 	} else {
 		Error := relaymodel.Error{
@@ -142,3 +183,30 @@ func RetrieveModel(c *gin.Context) {
 		})
 	}
 }
+
+func GetUserAvailableModels(c *gin.Context) {
+	ctx := c.Request.Context()
+	id := c.GetInt("id")
+	userGroup, err := model.CacheGetUserGroup(id)
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": err.Error(),
+		})
+		return
+	}
+	models, err := model.CacheGetGroupModels(ctx, userGroup)
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": err.Error(),
+		})
+		return
+	}
+	c.JSON(http.StatusOK, gin.H{
+		"success": true,
+		"message": "",
+		"data":    models,
+	})
+	return
+}

controller/token.go

@@ -1,10 +1,12 @@
 package controller
 
 import (
+	"fmt"
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/common/helper"
+	"github.com/songquanpeng/one-api/common/network"
 	"github.com/songquanpeng/one-api/model"
 	"net/http"
 	"strconv"
@@ -16,7 +18,10 @@ func GetAllTokens(c *gin.Context) {
 	if p < 0 {
 		p = 0
 	}
-	tokens, err := model.GetAllUserTokens(userId, p*config.ItemsPerPage, config.ItemsPerPage)
+
+	order := c.Query("order")
+	tokens, err := model.GetAllUserTokens(userId, p*config.ItemsPerPage, config.ItemsPerPage, order)
+
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
@@ -101,6 +106,19 @@ func GetTokenStatus(c *gin.Context) {
 	})
 }
 
+func validateToken(c *gin.Context, token model.Token) error {
+	if len(token.Name) > 30 {
+		return fmt.Errorf("令牌名称过长")
+	}
+	if token.Subnet != nil && *token.Subnet != "" {
+		err := network.IsValidSubnets(*token.Subnet)
+		if err != nil {
+			return fmt.Errorf("无效的网段：%s", err.Error())
+		}
+	}
+	return nil
+}
+
 func AddToken(c *gin.Context) {
 	token := model.Token{}
 	err := c.ShouldBindJSON(&token)
@@ -111,13 +129,15 @@ func AddToken(c *gin.Context) {
 		})
 		return
 	}
-	if len(token.Name) > 30 {
+	err = validateToken(c, token)
+	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
-			"message": "令牌名称过长",
+			"message": fmt.Sprintf("参数错误：%s", err.Error()),
 		})
 		return
 	}
+
 	cleanToken := model.Token{
 		UserId:         c.GetInt("id"),
 		Name:           token.Name,
@@ -127,6 +147,8 @@ func AddToken(c *gin.Context) {
 		ExpiredTime:    token.ExpiredTime,
 		RemainQuota:    token.RemainQuota,
 		UnlimitedQuota: token.UnlimitedQuota,
+		Models:         token.Models,
+		Subnet:         token.Subnet,
 	}
 	err = cleanToken.Insert()
 	if err != nil {
@@ -139,6 +161,7 @@ func AddToken(c *gin.Context) {
 	c.JSON(http.StatusOK, gin.H{
 		"success": true,
 		"message": "",
+		"data":    cleanToken,
 	})
 	return
 }
@@ -173,10 +196,11 @@ func UpdateToken(c *gin.Context) {
 		})
 		return
 	}
-	if len(token.Name) > 30 {
+	err = validateToken(c, token)
+	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
-			"message": "令牌名称过长",
+			"message": fmt.Sprintf("参数错误：%s", err.Error()),
 		})
 		return
 	}
@@ -212,6 +236,8 @@ func UpdateToken(c *gin.Context) {
 		cleanToken.ExpiredTime = token.ExpiredTime
 		cleanToken.RemainQuota = token.RemainQuota
 		cleanToken.UnlimitedQuota = token.UnlimitedQuota
+		cleanToken.Models = token.Models
+		cleanToken.Subnet = token.Subnet
 	}
 	err = cleanToken.Update()
 	if err != nil {

controller/user.go

@@ -58,11 +58,11 @@ func Login(c *gin.Context) {
 		})
 		return
 	}
-	setupLogin(&user, c)
+	SetupLogin(&user, c)
 }
 
 // setup session & cookies and then return user info
-func setupLogin(user *model.User, c *gin.Context) {
+func SetupLogin(user *model.User, c *gin.Context) {
 	session := sessions.Default(c)
 	session.Set("id", user.Id)
 	session.Set("username", user.Username)
@@ -184,7 +184,10 @@ func GetAllUsers(c *gin.Context) {
 	if p < 0 {
 		p = 0
 	}
-	users, err := model.GetAllUsers(p*config.ItemsPerPage, config.ItemsPerPage)
+
+	order := c.DefaultQuery("order", "")
+	users, err := model.GetAllUsers(p*config.ItemsPerPage, config.ItemsPerPage, order)
+
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
@@ -192,12 +195,12 @@ func GetAllUsers(c *gin.Context) {
 		})
 		return
 	}
+
 	c.JSON(http.StatusOK, gin.H{
 		"success": true,
 		"message": "",
 		"data":    users,
 	})
-	return
 }
 
 func SearchUsers(c *gin.Context) {
@@ -767,3 +770,38 @@ func TopUp(c *gin.Context) {
 	})
 	return
 }
+
+type adminTopUpRequest struct {
+	UserId int    `json:"user_id"`
+	Quota  int    `json:"quota"`
+	Remark string `json:"remark"`
+}
+
+func AdminTopUp(c *gin.Context) {
+	req := adminTopUpRequest{}
+	err := c.ShouldBindJSON(&req)
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": err.Error(),
+		})
+		return
+	}
+	err = model.IncreaseUserQuota(req.UserId, int64(req.Quota))
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": err.Error(),
+		})
+		return
+	}
+	if req.Remark == "" {
+		req.Remark = fmt.Sprintf("通过 API 充值 %s", common.LogQuota(int64(req.Quota)))
+	}
+	model.RecordTopupLog(req.UserId, req.Remark, req.Quota)
+	c.JSON(http.StatusOK, gin.H{
+		"success": true,
+		"message": "",
+	})
+	return
+}

docs/API.md

@@ -0,0 +1,53 @@
+# 使用 API 操控 & 扩展 One API
+> 欢迎提交 PR 在此放上你的拓展项目。
+
+例如，虽然 One API 本身没有直接支持支付，但是你可以通过系统扩展的 API 来实现支付功能。
+
+又或者你想自定义渠道管理策略，也可以通过 API 来实现渠道的禁用与启用。
+
+## 鉴权
+One API 支持两种鉴权方式：Cookie 和 Token，对于 Token，参照下图获取：
+
+![image](https://github.com/songquanpeng/songquanpeng.github.io/assets/39998050/c15281a7-83ed-47cb-a1f6-913cb6bf4a7c)
+
+之后，将 Token 作为请求头的 Authorization 字段的值即可，例如下面使用 Token 调用测试渠道的 API：
+![image](https://github.com/songquanpeng/songquanpeng.github.io/assets/39998050/1273b7ae-cb60-4c0d-93a6-b1cbc039c4f8)
+
+## 请求格式与响应格式
+One API 使用 JSON 格式进行请求和响应。
+
+对于响应体，一般格式如下：
+```json
+{
+  "message": "请求信息",
+  "success": true,
+  "data": {}
+}
+```
+
+## API 列表
+> 当前 API 列表不全，请自行通过浏览器抓取前端请求
+
+如果现有的 API 没有办法满足你的需求，欢迎提交 issue 讨论。
+
+### 获取当前登录用户信息
+**GET** `/api/user/self`
+
+### 为给定用户充值额度
+**POST** `/api/topup`
+```json
+{
+  "user_id": 1,
+  "quota": 100000,
+  "remark": "充值 100000 额度"
+}
+```
+
+## 其他
+### 充值链接上的附加参数
+One API 会在用户点击充值按钮的时候，将用户的信息和充值信息附加在链接上，例如：
+`https://example.com?username=root&user_id=1&transaction_id=4b3eed80-55d5-443f-bd44-fb18c648c837`
+
+你可以通过解析链接上的参数来获取用户信息和充值信息，然后调用 API 来为用户充值。
+
+注意，不是所有主题都支持该功能，欢迎 PR 补齐。

go.mod

@@ -15,6 +15,7 @@ require (
 	github.com/google/uuid v1.3.0
 	github.com/gorilla/websocket v1.5.0
 	github.com/pkoukk/tiktoken-go v0.1.5
+	github.com/smartystreets/goconvey v1.8.1
 	github.com/stretchr/testify v1.8.3
 	golang.org/x/crypto v0.17.0
 	golang.org/x/image v0.14.0
@@ -37,15 +38,18 @@ require (
 	github.com/go-playground/universal-translator v0.18.1 // indirect
 	github.com/go-sql-driver/mysql v1.6.0 // indirect
 	github.com/goccy/go-json v0.10.2 // indirect
+	github.com/gopherjs/gopherjs v1.17.2 // indirect
 	github.com/gorilla/context v1.1.1 // indirect
 	github.com/gorilla/securecookie v1.1.1 // indirect
 	github.com/gorilla/sessions v1.2.1 // indirect
 	github.com/jackc/pgpassfile v1.0.0 // indirect
 	github.com/jackc/pgservicefile v0.0.0-20221227161230-091c0ba34f0a // indirect
-	github.com/jackc/pgx/v5 v5.3.1 // indirect
+	github.com/jackc/pgx/v5 v5.5.4 // indirect
+	github.com/jackc/puddle/v2 v2.2.1 // indirect
 	github.com/jinzhu/inflection v1.0.0 // indirect
 	github.com/jinzhu/now v1.1.5 // indirect
 	github.com/json-iterator/go v1.1.12 // indirect
+	github.com/jtolds/gls v4.20.0+incompatible // indirect
 	github.com/klauspost/cpuid/v2 v2.2.4 // indirect
 	github.com/leodido/go-urn v1.2.4 // indirect
 	github.com/mattn/go-isatty v0.0.19 // indirect
@@ -54,10 +58,12 @@ require (
 	github.com/modern-go/reflect2 v1.0.2 // indirect
 	github.com/pelletier/go-toml/v2 v2.0.8 // indirect
 	github.com/pmezard/go-difflib v1.0.0 // indirect
+	github.com/smarty/assertions v1.15.0 // indirect
 	github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
 	github.com/ugorji/go/codec v1.2.11 // indirect
 	golang.org/x/arch v0.3.0 // indirect
 	golang.org/x/net v0.17.0 // indirect
+	golang.org/x/sync v0.1.0 // indirect
 	golang.org/x/sys v0.15.0 // indirect
 	golang.org/x/text v0.14.0 // indirect
 	google.golang.org/protobuf v1.33.0 // indirect

go.sum

@@ -56,11 +56,13 @@ github.com/golang-jwt/jwt v3.2.2+incompatible h1:IfV12K8xAKAnZqdXVzCZ+TOjboZ2keL
 github.com/golang-jwt/jwt v3.2.2+incompatible/go.mod h1:8pz2t5EyA70fFQQSrl6XZXzqecmYZeUEB8OUGHkxJ+I=
 github.com/golang/protobuf v1.3.3/go.mod h1:vzj43D7+SQXF/4pzW/hwtAqwc6iTitCiVSaWz5lYuqw=
 github.com/golang/protobuf v1.5.0/go.mod h1:FsONVRAS9T7sI+LIUmWTfcYkHO4aIWwzhcaSAoJOfIk=
-github.com/google/go-cmp v0.5.5 h1:Khx7svrCpmxxtHBq5j2mp/xVjsi8hQMfNLvJFAlrGgU=
 github.com/google/go-cmp v0.5.5/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
+github.com/google/go-cmp v0.5.6 h1:BKbKCqvP6I+rmFHt06ZmyQtvB8xAkWdhFyr0ZUNZcxQ=
 github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
 github.com/google/uuid v1.3.0 h1:t6JiXgmwXMjEs8VusXIJk2BXHsn+wx8BZdTaoZ5fu7I=
 github.com/google/uuid v1.3.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
+github.com/gopherjs/gopherjs v1.17.2 h1:fQnZVsXk8uxXIStYb0N4bGk7jeyTalG/wsZjQ25dO0g=
+github.com/gopherjs/gopherjs v1.17.2/go.mod h1:pRRIvn/QzFLrKfvEz3qUuEhtE/zLCWfreZ6J5gM2i+k=
 github.com/gorilla/context v1.1.1 h1:AWwleXJkX/nhcU9bZSnZoi3h/qGYqQAGhq6zZe/aQW8=
 github.com/gorilla/context v1.1.1/go.mod h1:kBGZzfjB9CEq2AlWe17Uuf7NDRt0dE0s8S51q0aT7Yg=
 github.com/gorilla/securecookie v1.1.1 h1:miw7JPhV+b/lAHSXz4qd/nN9jRiAFV5FwjeKyCS8BvQ=
@@ -73,8 +75,10 @@ github.com/jackc/pgpassfile v1.0.0 h1:/6Hmqy13Ss2zCq62VdNG8tM1wchn8zjSGOBJ6icpsI
 github.com/jackc/pgpassfile v1.0.0/go.mod h1:CEx0iS5ambNFdcRtxPj5JhEz+xB6uRky5eyVu/W2HEg=
 github.com/jackc/pgservicefile v0.0.0-20221227161230-091c0ba34f0a h1:bbPeKD0xmW/Y25WS6cokEszi5g+S0QxI/d45PkRi7Nk=
 github.com/jackc/pgservicefile v0.0.0-20221227161230-091c0ba34f0a/go.mod h1:5TJZWKEWniPve33vlWYSoGYefn3gLQRzjfDlhSJ9ZKM=
-github.com/jackc/pgx/v5 v5.3.1 h1:Fcr8QJ1ZeLi5zsPZqQeUZhNhxfkkKBOgJuYkJHoBOtU=
-github.com/jackc/pgx/v5 v5.3.1/go.mod h1:t3JDKnCBlYIc0ewLF0Q7B8MXmoIaBOZj/ic7iHozM/8=
+github.com/jackc/pgx/v5 v5.5.4 h1:Xp2aQS8uXButQdnCMWNmvx6UysWQQC+u1EoizjguY+8=
+github.com/jackc/pgx/v5 v5.5.4/go.mod h1:ez9gk+OAat140fv9ErkZDYFWmXLfV+++K0uAOiwgm1A=
+github.com/jackc/puddle/v2 v2.2.1 h1:RhxXJtFG022u4ibrCSMSiu5aOq1i77R3OHKNJj77OAk=
+github.com/jackc/puddle/v2 v2.2.1/go.mod h1:vriiEXHvEE654aYKXXjOvZM39qJ0q+azkZFrfEOc3H4=
 github.com/jinzhu/inflection v1.0.0 h1:K317FqzuhWc8YvSVlFMCCUb36O/S9MCKRDI7QkRKD/E=
 github.com/jinzhu/inflection v1.0.0/go.mod h1:h+uFLlag+Qp1Va5pdKtLDYj+kHp5pxUVkryuEj+Srlc=
 github.com/jinzhu/now v1.1.4/go.mod h1:d3SSVoowX0Lcu0IBviAWJpolVfI5UJVZZ7cO71lE/z8=
@@ -83,6 +87,8 @@ github.com/jinzhu/now v1.1.5/go.mod h1:d3SSVoowX0Lcu0IBviAWJpolVfI5UJVZZ7cO71lE/
 github.com/json-iterator/go v1.1.9/go.mod h1:KdQUCv79m/52Kvf8AW2vK1V8akMuk1QjK/uOdHXbAo4=
 github.com/json-iterator/go v1.1.12 h1:PV8peI4a0ysnczrg+LtxykD8LfKY9ML6u2jnxaEnrnM=
 github.com/json-iterator/go v1.1.12/go.mod h1:e30LSqwooZae/UwlEbR2852Gd8hjQvJoHmT4TnhNGBo=
+github.com/jtolds/gls v4.20.0+incompatible h1:xdiiI2gbIgH/gLH7ADydsJ1uDOEzR8yvV7C0MuV77Wo=
+github.com/jtolds/gls v4.20.0+incompatible/go.mod h1:QJZ7F/aHp+rZTRtaJ1ow/lLfFfVYBRgL+9YlvaHOwJU=
 github.com/klauspost/cpuid/v2 v2.0.9/go.mod h1:FInQzS24/EEf25PyTYn52gqo7WaD8xa0213Md/qVLRg=
 github.com/klauspost/cpuid/v2 v2.2.4 h1:acbojRNwl3o09bUq+yDCtZFc1aiwaAAxtcn8YkZXnvk=
 github.com/klauspost/cpuid/v2 v2.2.4/go.mod h1:RVVoqg1df56z8g3pUjL/3lE5UfnlrJX8tyFgg4nqhuY=
@@ -125,6 +131,10 @@ github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZN
 github.com/rogpeppe/go-internal v1.6.1/go.mod h1:xXDCJY+GAPziupqXw64V24skbSoqbTEfhy4qGm1nDQc=
 github.com/rogpeppe/go-internal v1.8.0 h1:FCbCCtXNOY3UtUuHUYaghJg4y7Fd14rXifAYUAtL9R8=
 github.com/rogpeppe/go-internal v1.8.0/go.mod h1:WmiCO8CzOY8rg0OYDC4/i/2WRWAB6poM+XZ2dLUbcbE=
+github.com/smarty/assertions v1.15.0 h1:cR//PqUBUiQRakZWqBiFFQ9wb8emQGDb0HeGdqGByCY=
+github.com/smarty/assertions v1.15.0/go.mod h1:yABtdzeQs6l1brC900WlRNwj6ZR55d7B+E8C6HtKdec=
+github.com/smartystreets/goconvey v1.8.1 h1:qGjIddxOk4grTu9JPOU31tVfq3cNdBlNa5sSznIX1xY=
+github.com/smartystreets/goconvey v1.8.1/go.mod h1:+/u4qLyY6x1jReYOp7GOM2FSt8aP9CzCZL03bI28W60=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
 github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=
@@ -157,6 +167,8 @@ golang.org/x/image v0.14.0/go.mod h1:HUYqC05R2ZcZ3ejNQsIHQDQiwWM4JBqmm6MKANTp4LE
 golang.org/x/net v0.0.0-20210226172049-e18ecbb05110/go.mod h1:m0MpNAwzfU5UDzcl9v0D8zg8gWTRqZa9RBIspLL5mdg=
 golang.org/x/net v0.17.0 h1:pVaXccu2ozPjCXewfr1S7xza/zcXTity9cCdXQYSjIM=
 golang.org/x/net v0.17.0/go.mod h1:NxSsAGuq816PNPmqtQdLE42eU2Fs7NoRIZrHJAlaCOE=
+golang.org/x/sync v0.1.0 h1:wsuoTGHzEhffawBOhz5CYhcrV4IdKZbEyZjBMuTp12o=
+golang.org/x/sync v0.1.0/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sys v0.0.0-20200116001909-b77594299b42/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20201119102817-f84b799fce68/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20210615035016-665e8c7367d1/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
@@ -173,8 +185,8 @@ golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.14.0 h1:ScX5w1eTa3QqT8oi6+ziP7dTV1S2+ALU0bI+0zXKWiQ=
 golang.org/x/text v0.14.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
-golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543 h1:E7g+9GITq07hpfrRu66IVDexMakfv52eLZ2CXBWiKr4=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
+golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1 h1:go1bK/D/BFZV2I8cIQd1NKEZ+0owSTG1fDTci4IqFcE=
 google.golang.org/protobuf v1.26.0-rc.1/go.mod h1:jlhhOSvTdKEhbULTjvd4ARK9grFBp09yW+WbY/TyQbw=
 google.golang.org/protobuf v1.28.0/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqwMG9pJV4I=
 google.golang.org/protobuf v1.33.0 h1:uNO2rsAINq/JlFpSdYEKIZ0uKD/R9cpdv0T+yoGwGmI=

i18n/en.json

@@ -8,12 +8,12 @@
   "确认删除": "Confirm Delete",
   "确认绑定": "Confirm Binding",
   "您正在删除自己的帐户，将清空所有数据且不可恢复": "You are deleting your account, all data will be cleared and unrecoverable.",
-  "\"通道「%s」（#%d）已被禁用\"": "\"Channel %s (#%d) has been disabled\"",
-  "通道「%s」（#%d）已被禁用，原因：%s": "Channel %s (#%d) has been disabled, reason: %s",
+  "\"渠道「%s」（#%d）已被禁用\"": "\"Channel %s (#%d) has been disabled\"",
+  "渠道「%s」（#%d）已被禁用，原因：%s": "Channel %s (#%d) has been disabled, reason: %s",
   "测试已在运行中": "Test is already running",
   "响应时间 %.2fs 超过阈值 %.2fs": "Response time %.2fs exceeds threshold %.2fs",
-  "通道测试完成": "Channel test completed",
-  "通道测试完成，如果没有收到禁用通知，说明所有通道都正常": "Channel test completed, if you have not received the disable notification, it means that all channels are normal",
+  "渠道测试完成": "Channel test completed",
+  "渠道测试完成，如果没有收到禁用通知，说明所有渠道都正常": "Channel test completed, if you have not received the disable notification, it means that all channels are normal",
   "无法连接至 GitHub 服务器，请稍后重试！": "Unable to connect to GitHub server, please try again later!",
   "返回值非法，用户字段为空，请稍后重试！": "The return value is illegal, the user field is empty, please try again later!",
   "管理员未开启通过 GitHub 登录以及注册": "The administrator did not turn on login and registration via GitHub",
@@ -119,11 +119,11 @@
   " 个月 ": " M ",
   " 年 ": " y ",
   "未测试": "Not tested",
-  "通道 ${name} 测试成功，耗时 ${time.toFixed(2)} 秒。": "Channel ${name} test succeeded, time consumed ${time.toFixed(2)} s.",
-  "已成功开始测试所有通道，请刷新页面查看结果。": "All channels have been successfully tested, please refresh the page to view the results.",
-  "已成功开始测试所有已启用通道，请刷新页面查看结果。": "All enabled channels have been successfully tested, please refresh the page to view the results.",
-  "通道 ${name} 余额更新成功！": "Channel ${name} balance updated successfully!",
-  "已更新完毕所有已启用通道余额！": "The balance of all enabled channels has been updated!",
+  "渠道 ${name} 测试成功，耗时 ${time.toFixed(2)} 秒。": "Channel ${name} test succeeded, time consumed ${time.toFixed(2)} s.",
+  "已成功开始测试所有渠道，请刷新页面查看结果。": "All channels have been successfully tested, please refresh the page to view the results.",
+  "已成功开始测试所有已启用渠道，请刷新页面查看结果。": "All enabled channels have been successfully tested, please refresh the page to view the results.",
+  "渠道 ${name} 余额更新成功！": "Channel ${name} balance updated successfully!",
+  "已更新完毕所有已启用渠道余额！": "The balance of all enabled channels has been updated!",
   "搜索渠道的 ID，名称和密钥 ...": "Search for channel ID, name and key ...",
   "名称": "Name",
   "分组": "Group",
@@ -141,9 +141,9 @@
   "启用": "Enable",
   "编辑": "Edit",
   "添加新的渠道": "Add a new channel",
-  "测试所有通道": "Test all channels",
-  "测试所有已启用通道": "Test all enabled channels",
-  "更新所有已启用通道余额": "Update the balance of all enabled channels",
+  "测试所有渠道": "Test all channels",
+  "测试所有已启用渠道": "Test all enabled channels",
+  "更新所有已启用渠道余额": "Update the balance of all enabled channels",
   "刷新": "Refresh",
   "处理中...": "Processing...",
   "绑定成功！": "Binding succeeded!",
@@ -207,11 +207,11 @@
   "监控设置": "Monitoring Settings",
   "最长响应时间": "Longest Response Time",
   "单位秒": "Unit in seconds",
-  "当运行通道全部测试时": "When all operating channels are tested",
-  "超过此时间将自动禁用通道": "Channels will be automatically disabled if this time is exceeded",
+  "当运行渠道全部测试时": "When all operating channels are tested",
+  "超过此时间将自动禁用渠道": "Channels will be automatically disabled if this time is exceeded",
   "额度提醒阈值": "Quota reminder threshold",
   "低于此额度时将发送邮件提醒用户": "Email will be sent to remind users when the quota is below this",
-  "失败时自动禁用通道": "Automatically disable the channel when it fails",
+  "失败时自动禁用渠道": "Automatically disable the channel when it fails",
   "保存监控设置": "Save Monitoring Settings",
   "额度设置": "Quota Settings",
   "新用户初始额度": "Initial quota for new users",
@@ -405,7 +405,7 @@
   "镜像": "Mirror",
   "请输入镜像站地址，格式为：https://domain.com，可不填，不填则使用渠道默认值": "Please enter the mirror site address, the format is: https://domain.com, it can be left blank, if left blank, the default value of the channel will be used",
   "模型": "Model",
-  "请选择该通道所支持的模型": "Please select the model supported by the channel",
+  "请选择该渠道所支持的模型": "Please select the model supported by the channel",
   "填入基础模型": "Fill in the basic model",
   "填入所有模型": "Fill in all models",
   "清除所有模型": "Clear all models",
@@ -515,7 +515,7 @@
   "请输入自定义渠道的 Base URL": "Please enter the Base URL of the custom channel",
   "Homepage URL 填": "Fill in the Homepage URL",
   "Authorization callback URL 填": "Fill in the Authorization callback URL",
-  "请为通道命名": "Please name the channel",
+  "请为渠道命名": "Please name the channel",
   "此项可选，用于修改请求体中的模型名称，为一个 JSON 字符串，键为请求中模型名称，值为要替换的模型名称，例如：": "This is optional, used to modify the model name in the request body, it's a JSON string, the key is the model name in the request, and the value is the model name to be replaced, for example:",
   "模型重定向": "Model redirection",
   "请输入渠道对应的鉴权密钥": "Please enter the authentication key corresponding to the channel",

middleware/auth.go

@@ -1,10 +1,12 @@
 package middleware
 
 import (
+	"fmt"
 	"github.com/gin-contrib/sessions"
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/blacklist"
+	"github.com/songquanpeng/one-api/common/network"
 	"github.com/songquanpeng/one-api/model"
 	"net/http"
 	"strings"
@@ -88,6 +90,7 @@ func RootAuth() func(c *gin.Context) {
 
 func TokenAuth() func(c *gin.Context) {
 	return func(c *gin.Context) {
+		ctx := c.Request.Context()
 		key := c.Request.Header.Get("Authorization")
 		key = strings.TrimPrefix(key, "Bearer ")
 		key = strings.TrimPrefix(key, "sk-")
@@ -98,6 +101,12 @@ func TokenAuth() func(c *gin.Context) {
 			abortWithMessage(c, http.StatusUnauthorized, err.Error())
 			return
 		}
+		if token.Subnet != nil && *token.Subnet != "" {
+			if !network.IsIpInSubnets(ctx, c.ClientIP(), *token.Subnet) {
+				abortWithMessage(c, http.StatusForbidden, fmt.Sprintf("该令牌只能在指定网段使用：%s，当前 ip：%s", *token.Subnet, c.ClientIP()))
+				return
+			}
+		}
 		userEnabled, err := model.CacheIsUserEnabled(token.UserId)
 		if err != nil {
 			abortWithMessage(c, http.StatusInternalServerError, err.Error())
@@ -107,6 +116,19 @@ func TokenAuth() func(c *gin.Context) {
 			abortWithMessage(c, http.StatusForbidden, "用户已被封禁")
 			return
 		}
+		requestModel, err := getRequestModel(c)
+		if err != nil && !strings.HasPrefix(c.Request.URL.Path, "/v1/models") {
+			abortWithMessage(c, http.StatusBadRequest, err.Error())
+			return
+		}
+		c.Set("request_model", requestModel)
+		if token.Models != nil && *token.Models != "" {
+			c.Set("available_models", *token.Models)
+			if requestModel != "" && !isModelInList(requestModel, *token.Models) {
+				abortWithMessage(c, http.StatusForbidden, fmt.Sprintf("该令牌无权使用模型：%s", requestModel))
+				return
+			}
+		}
 		c.Set("id", token.UserId)
 		c.Set("token_id", token.Id)
 		c.Set("token_name", token.Name)

middleware/distributor.go

@@ -2,14 +2,12 @@ package middleware
 
 import (
 	"fmt"
+	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/model"
 	"net/http"
 	"strconv"
-	"strings"
-
-	"github.com/gin-gonic/gin"
 )
 
 type ModelRequest struct {
@@ -40,37 +38,11 @@ func Distribute() func(c *gin.Context) {
 				return
 			}
 		} else {
-			// Select a channel for the user
-			var modelRequest ModelRequest
-			err := common.UnmarshalBodyReusable(c, &modelRequest)
+			requestModel := c.GetString("request_model")
+			var err error
+			channel, err = model.CacheGetRandomSatisfiedChannel(userGroup, requestModel, false)
 			if err != nil {
-				abortWithMessage(c, http.StatusBadRequest, "无效的请求")
-				return
-			}
-			if strings.HasPrefix(c.Request.URL.Path, "/v1/moderations") {
-				if modelRequest.Model == "" {
-					modelRequest.Model = "text-moderation-stable"
-				}
-			}
-			if strings.HasSuffix(c.Request.URL.Path, "embeddings") {
-				if modelRequest.Model == "" {
-					modelRequest.Model = c.Param("model")
-				}
-			}
-			if strings.HasPrefix(c.Request.URL.Path, "/v1/images/generations") {
-				if modelRequest.Model == "" {
-					modelRequest.Model = "dall-e-2"
-				}
-			}
-			if strings.HasPrefix(c.Request.URL.Path, "/v1/audio/transcriptions") || strings.HasPrefix(c.Request.URL.Path, "/v1/audio/translations") {
-				if modelRequest.Model == "" {
-					modelRequest.Model = "whisper-1"
-				}
-			}
-			requestModel = modelRequest.Model
-			channel, err = model.CacheGetRandomSatisfiedChannel(userGroup, modelRequest.Model, false)
-			if err != nil {
-				message := fmt.Sprintf("当前分组 %s 下对于模型 %s 无可用渠道", userGroup, modelRequest.Model)
+				message := fmt.Sprintf("当前分组 %s 下对于模型 %s 无可用渠道", userGroup, requestModel)
 				if channel != nil {
 					logger.SysError(fmt.Sprintf("渠道不存在：%d", channel.Id))
 					message = "数据库一致性已被破坏，请联系管理员"

middleware/utils.go

@@ -1,9 +1,12 @@
 package middleware
 
 import (
+	"fmt"
 	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/logger"
+	"strings"
 )
 
 func abortWithMessage(c *gin.Context, statusCode int, message string) {
@@ -16,3 +19,42 @@ func abortWithMessage(c *gin.Context, statusCode int, message string) {
 	c.Abort()
 	logger.Error(c.Request.Context(), message)
 }
+
+func getRequestModel(c *gin.Context) (string, error) {
+	var modelRequest ModelRequest
+	err := common.UnmarshalBodyReusable(c, &modelRequest)
+	if err != nil {
+		return "", fmt.Errorf("common.UnmarshalBodyReusable failed: %w", err)
+	}
+	if strings.HasPrefix(c.Request.URL.Path, "/v1/moderations") {
+		if modelRequest.Model == "" {
+			modelRequest.Model = "text-moderation-stable"
+		}
+	}
+	if strings.HasSuffix(c.Request.URL.Path, "embeddings") {
+		if modelRequest.Model == "" {
+			modelRequest.Model = c.Param("model")
+		}
+	}
+	if strings.HasPrefix(c.Request.URL.Path, "/v1/images/generations") {
+		if modelRequest.Model == "" {
+			modelRequest.Model = "dall-e-2"
+		}
+	}
+	if strings.HasPrefix(c.Request.URL.Path, "/v1/audio/transcriptions") || strings.HasPrefix(c.Request.URL.Path, "/v1/audio/translations") {
+		if modelRequest.Model == "" {
+			modelRequest.Model = "whisper-1"
+		}
+	}
+	return modelRequest.Model, nil
+}
+
+func isModelInList(modelName string, models string) bool {
+	modelList := strings.Split(models, ",")
+	for _, model := range modelList {
+		if modelName == model {
+			return true
+		}
+	}
+	return false
+}

model/ability.go

@@ -1,7 +1,10 @@
 package model
 
 import (
+	"context"
 	"github.com/songquanpeng/one-api/common"
+	"gorm.io/gorm"
+	"sort"
 	"strings"
 )
 
@@ -13,7 +16,7 @@ type Ability struct {
 	Priority  *int64 `json:"priority" gorm:"bigint;default:0;index"`
 }
 
-func GetRandomSatisfiedChannel(group string, model string) (*Channel, error) {
+func GetRandomSatisfiedChannel(group string, model string, ignoreFirstPriority bool) (*Channel, error) {
 	ability := Ability{}
 	groupCol := "`group`"
 	trueVal := "1"
@@ -23,8 +26,13 @@ func GetRandomSatisfiedChannel(group string, model string) (*Channel, error) {
 	}
 
 	var err error = nil
-	maxPrioritySubQuery := DB.Model(&Ability{}).Select("MAX(priority)").Where(groupCol+" = ? and model = ? and enabled = "+trueVal, group, model)
-	channelQuery := DB.Where(groupCol+" = ? and model = ? and enabled = "+trueVal+" and priority = (?)", group, model, maxPrioritySubQuery)
+	var channelQuery *gorm.DB
+	if ignoreFirstPriority {
+		channelQuery = DB.Where(groupCol+" = ? and model = ? and enabled = "+trueVal, group, model)
+	} else {
+		maxPrioritySubQuery := DB.Model(&Ability{}).Select("MAX(priority)").Where(groupCol+" = ? and model = ? and enabled = "+trueVal, group, model)
+		channelQuery = DB.Where(groupCol+" = ? and model = ? and enabled = "+trueVal+" and priority = (?)", group, model, maxPrioritySubQuery)
+	}
 	if common.UsingSQLite || common.UsingPostgreSQL {
 		err = channelQuery.Order("RANDOM()").First(&ability).Error
 	} else {
@@ -82,3 +90,19 @@ func (channel *Channel) UpdateAbilities() error {
 func UpdateAbilityStatus(channelId int, status bool) error {
 	return DB.Model(&Ability{}).Where("channel_id = ?", channelId).Select("enabled").Update("enabled", status).Error
 }
+
+func GetGroupModels(ctx context.Context, group string) ([]string, error) {
+	groupCol := "`group`"
+	trueVal := "1"
+	if common.UsingPostgreSQL {
+		groupCol = `"group"`
+		trueVal = "true"
+	}
+	var models []string
+	err := DB.Model(&Ability{}).Distinct("model").Where(groupCol+" = ? and enabled = "+trueVal, group).Pluck("model", &models).Error
+	if err != nil {
+		return nil, err
+	}
+	sort.Strings(models)
+	return models, err
+}

model/cache.go

@@ -21,6 +21,7 @@ var (
 	UserId2GroupCacheSeconds  = config.SyncFrequency
 	UserId2QuotaCacheSeconds  = config.SyncFrequency
 	UserId2StatusCacheSeconds = config.SyncFrequency
+	GroupModelsCacheSeconds   = config.SyncFrequency
 )
 
 func CacheGetTokenByKey(key string) (*Token, error) {
@@ -146,6 +147,25 @@ func CacheIsUserEnabled(userId int) (bool, error) {
 	return userEnabled, err
 }
 
+func CacheGetGroupModels(ctx context.Context, group string) ([]string, error) {
+	if !common.RedisEnabled {
+		return GetGroupModels(ctx, group)
+	}
+	modelsStr, err := common.RedisGet(fmt.Sprintf("group_models:%s", group))
+	if err == nil {
+		return strings.Split(modelsStr, ","), nil
+	}
+	models, err := GetGroupModels(ctx, group)
+	if err != nil {
+		return nil, err
+	}
+	err = common.RedisSet(fmt.Sprintf("group_models:%s", group), strings.Join(models, ","), time.Duration(GroupModelsCacheSeconds)*time.Second)
+	if err != nil {
+		logger.SysError("Redis set group models error: " + err.Error())
+	}
+	return models, nil
+}
+
 var group2model2channels map[string]map[string][]*Channel
 var channelSyncLock sync.RWMutex
 
@@ -205,7 +225,7 @@ func SyncChannelCache(frequency int) {
 
 func CacheGetRandomSatisfiedChannel(group string, model string, ignoreFirstPriority bool) (*Channel, error) {
 	if !config.MemoryCacheEnabled {
-		return GetRandomSatisfiedChannel(group, model)
+		return GetRandomSatisfiedChannel(group, model, ignoreFirstPriority)
 	}
 	channelSyncLock.RLock()
 	defer channelSyncLock.RUnlock()

model/log.go

@@ -51,6 +51,21 @@ func RecordLog(userId int, logType int, content string) {
 	}
 }
 
+func RecordTopupLog(userId int, content string, quota int) {
+	log := &Log{
+		UserId:    userId,
+		Username:  GetUsernameById(userId),
+		CreatedAt: helper.GetTimestamp(),
+		Type:      LogTypeTopup,
+		Content:   content,
+		Quota:     quota,
+	}
+	err := LOG_DB.Create(log).Error
+	if err != nil {
+		logger.SysError("failed to record log: " + err.Error())
+	}
+}
+
 func RecordConsumeLog(ctx context.Context, userId int, channelId int, promptTokens int, completionTokens int, modelName string, tokenName string, quota int64, content string) {
 	logger.Info(ctx, fmt.Sprintf("record consume log: userId=%d, channelId=%d, promptTokens=%d, completionTokens=%d, modelName=%s, tokenName=%s, quota=%d, content=%s", userId, channelId, promptTokens, completionTokens, modelName, tokenName, quota, content))
 	if !config.LogConsumeEnabled {

model/main.go

@@ -23,7 +23,7 @@ func CreateRootAccountIfNeed() error {
 	var user User
 	//if user.Status != util.UserStatusEnabled {
 	if err := DB.First(&user).Error; err != nil {
-		logger.SysLog("no user exists, create a root user for you: username is root, password is 123456")
+		logger.SysLog("no user exists, creating a root user for you: username is root, password is 123456")
 		hashedPassword, err := common.Password2Hash("123456")
 		if err != nil {
 			return err
@@ -35,9 +35,25 @@ func CreateRootAccountIfNeed() error {
 			Status:      common.UserStatusEnabled,
 			DisplayName: "Root User",
 			AccessToken: helper.GetUUID(),
-			Quota:       100000000,
+			Quota:       500000000000000,
 		}
 		DB.Create(&rootUser)
+		if config.InitialRootToken != "" {
+			logger.SysLog("creating initial root token as requested")
+			token := Token{
+				Id:             1,
+				UserId:         rootUser.Id,
+				Key:            config.InitialRootToken,
+				Status:         common.TokenStatusEnabled,
+				Name:           "Initial Root Token",
+				CreatedTime:    helper.GetTimestamp(),
+				AccessedTime:   helper.GetTimestamp(),
+				ExpiredTime:    -1,
+				RemainQuota:    500000000000000,
+				UnlimitedQuota: true,
+			}
+			DB.Create(&token)
+		}
 	}
 	return nil
 }

model/option.go

@@ -172,6 +172,10 @@ func updateOptionMap(key string, value string) (err error) {
 		config.GitHubClientId = value
 	case "GitHubClientSecret":
 		config.GitHubClientSecret = value
+	case "LarkClientId":
+		config.LarkClientId = value
+	case "LarkClientSecret":
+		config.LarkClientSecret = value
 	case "Footer":
 		config.Footer = value
 	case "SystemName":

model/redemption.go

@@ -14,7 +14,7 @@ type Redemption struct {
 	Key          string `json:"key" gorm:"type:char(32);uniqueIndex"`
 	Status       int    `json:"status" gorm:"default:1"`
 	Name         string `json:"name" gorm:"index"`
-	Quota        int64  `json:"quota" gorm:"default:100"`
+	Quota        int64  `json:"quota" gorm:"bigint;default:100"`
 	CreatedTime  int64  `json:"created_time" gorm:"bigint"`
 	RedeemedTime int64  `json:"redeemed_time" gorm:"bigint"`
 	Count        int    `json:"count" gorm:"-:all"` // only for api request

model/token.go

@@ -12,23 +12,36 @@ import (
 )
 
 type Token struct {
-	Id             int    `json:"id"`
-	UserId         int    `json:"user_id"`
-	Key            string `json:"key" gorm:"type:char(48);uniqueIndex"`
-	Status         int    `json:"status" gorm:"default:1"`
-	Name           string `json:"name" gorm:"index" `
-	CreatedTime    int64  `json:"created_time" gorm:"bigint"`
-	AccessedTime   int64  `json:"accessed_time" gorm:"bigint"`
-	ExpiredTime    int64  `json:"expired_time" gorm:"bigint;default:-1"` // -1 means never expired
-	RemainQuota    int64  `json:"remain_quota" gorm:"default:0"`
-	UnlimitedQuota bool   `json:"unlimited_quota" gorm:"default:false"`
-	UsedQuota      int64  `json:"used_quota" gorm:"default:0"` // used quota
+	Id             int     `json:"id"`
+	UserId         int     `json:"user_id"`
+	Key            string  `json:"key" gorm:"type:char(48);uniqueIndex"`
+	Status         int     `json:"status" gorm:"default:1"`
+	Name           string  `json:"name" gorm:"index" `
+	CreatedTime    int64   `json:"created_time" gorm:"bigint"`
+	AccessedTime   int64   `json:"accessed_time" gorm:"bigint"`
+	ExpiredTime    int64   `json:"expired_time" gorm:"bigint;default:-1"` // -1 means never expired
+	RemainQuota    int64   `json:"remain_quota" gorm:"bigint;default:0"`
+	UnlimitedQuota bool    `json:"unlimited_quota" gorm:"default:false"`
+	UsedQuota      int64   `json:"used_quota" gorm:"bigint;default:0"` // used quota
+	Models         *string `json:"models" gorm:"default:''"`           // allowed models
+	Subnet         *string `json:"subnet" gorm:"default:''"`           // allowed subnet
 }
 
-func GetAllUserTokens(userId int, startIdx int, num int) ([]*Token, error) {
+func GetAllUserTokens(userId int, startIdx int, num int, order string) ([]*Token, error) {
 	var tokens []*Token
 	var err error
-	err = DB.Where("user_id = ?", userId).Order("id desc").Limit(num).Offset(startIdx).Find(&tokens).Error
+	query := DB.Where("user_id = ?", userId)
+
+	switch order {
+	case "remain_quota":
+		query = query.Order("unlimited_quota desc, remain_quota desc")
+	case "used_quota":
+		query = query.Order("used_quota desc")
+	default:
+		query = query.Order("id desc")
+	}
+
+	err = query.Limit(num).Offset(startIdx).Find(&tokens).Error
 	return tokens, err
 }
 
@@ -50,7 +63,7 @@ func ValidateUserToken(key string) (token *Token, err error) {
 		return nil, errors.New("令牌验证失败")
 	}
 	if token.Status == common.TokenStatusExhausted {
-		return nil, errors.New("该令牌额度已用尽")
+		return nil, fmt.Errorf("令牌 %s（#%d）额度已用尽", token.Name, token.Id)
 	} else if token.Status == common.TokenStatusExpired {
 		return nil, errors.New("该令牌已过期")
 	}
@@ -110,7 +123,7 @@ func (token *Token) Insert() error {
 // Update Make sure your token's fields is completed, because this will update non-zero values
 func (token *Token) Update() error {
 	var err error
-	err = DB.Model(token).Select("name", "status", "expired_time", "remain_quota", "unlimited_quota").Updates(token).Error
+	err = DB.Model(token).Select("name", "status", "expired_time", "remain_quota", "unlimited_quota", "models", "subnet").Updates(token).Error
 	return err
 }
 

model/user.go

@@ -24,11 +24,12 @@ type User struct {
 	Email            string `json:"email" gorm:"index" validate:"max=50"`
 	GitHubId         string `json:"github_id" gorm:"column:github_id;index"`
 	WeChatId         string `json:"wechat_id" gorm:"column:wechat_id;index"`
+	LarkId           string `json:"lark_id" gorm:"column:lark_id;index"`
 	VerificationCode string `json:"verification_code" gorm:"-:all"`                                    // this field is only for Email verification, don't save it to database!
 	AccessToken      string `json:"access_token" gorm:"type:char(32);column:access_token;uniqueIndex"` // this token is for system management
-	Quota            int64  `json:"quota" gorm:"type:int;default:0"`
-	UsedQuota        int64  `json:"used_quota" gorm:"type:int;default:0;column:used_quota"` // used quota
-	RequestCount     int    `json:"request_count" gorm:"type:int;default:0;"`               // request number
+	Quota            int64  `json:"quota" gorm:"bigint;default:0"`
+	UsedQuota        int64  `json:"used_quota" gorm:"bigint;default:0;column:used_quota"` // used quota
+	RequestCount     int    `json:"request_count" gorm:"type:int;default:0;"`             // request number
 	Group            string `json:"group" gorm:"type:varchar(32);default:'default'"`
 	AffCode          string `json:"aff_code" gorm:"type:varchar(32);column:aff_code;uniqueIndex"`
 	InviterId        int    `json:"inviter_id" gorm:"type:int;column:inviter_id;index"`
@@ -40,8 +41,21 @@ func GetMaxUserId() int {
 	return user.Id
 }
 
-func GetAllUsers(startIdx int, num int) (users []*User, err error) {
-	err = DB.Order("id desc").Limit(num).Offset(startIdx).Omit("password").Where("status != ?", common.UserStatusDeleted).Find(&users).Error
+func GetAllUsers(startIdx int, num int, order string) (users []*User, err error) {
+	query := DB.Limit(num).Offset(startIdx).Omit("password").Where("status != ?", common.UserStatusDeleted)
+
+	switch order {
+	case "quota":
+		query = query.Order("quota desc")
+	case "used_quota":
+		query = query.Order("used_quota desc")
+	case "request_count":
+		query = query.Order("request_count desc")
+	default:
+		query = query.Order("id desc")
+	}
+
+	err = query.Find(&users).Error
 	return users, err
 }
 
@@ -193,6 +207,14 @@ func (user *User) FillUserByGitHubId() error {
 	return nil
 }
 
+func (user *User) FillUserByLarkId() error {
+	if user.LarkId == "" {
+		return errors.New("lark id 为空！")
+	}
+	DB.Where(User{LarkId: user.LarkId}).First(user)
+	return nil
+}
+
 func (user *User) FillUserByWeChatId() error {
 	if user.WeChatId == "" {
 		return errors.New("WeChat id 为空！")
@@ -221,6 +243,10 @@ func IsGitHubIdAlreadyTaken(githubId string) bool {
 	return DB.Where("github_id = ?", githubId).Find(&User{}).RowsAffected == 1
 }
 
+func IsLarkIdAlreadyTaken(githubId string) bool {
+	return DB.Where("lark_id = ?", githubId).Find(&User{}).RowsAffected == 1
+}
+
 func IsUsernameAlreadyTaken(username string) bool {
 	return DB.Where("username = ?", username).Find(&User{}).RowsAffected == 1
 }

monitor/channel.go

@@ -31,17 +31,17 @@ func notifyRootUser(subject string, content string) {
 func DisableChannel(channelId int, channelName string, reason string) {
 	model.UpdateChannelStatusById(channelId, common.ChannelStatusAutoDisabled)
 	logger.SysLog(fmt.Sprintf("channel #%d has been disabled: %s", channelId, reason))
-	subject := fmt.Sprintf("通道「%s」（#%d）已被禁用", channelName, channelId)
-	content := fmt.Sprintf("通道「%s」（#%d）已被禁用，原因：%s", channelName, channelId, reason)
+	subject := fmt.Sprintf("渠道「%s」（#%d）已被禁用", channelName, channelId)
+	content := fmt.Sprintf("渠道「%s」（#%d）已被禁用，原因：%s", channelName, channelId, reason)
 	notifyRootUser(subject, content)
 }
 
 func MetricDisableChannel(channelId int, successRate float64) {
 	model.UpdateChannelStatusById(channelId, common.ChannelStatusAutoDisabled)
 	logger.SysLog(fmt.Sprintf("channel #%d has been disabled due to low success rate: %.2f", channelId, successRate*100))
-	subject := fmt.Sprintf("通道 #%d 已被禁用", channelId)
-	content := fmt.Sprintf("该渠道在最近 %d 次调用中成功率为 %.2f%%，低于阈值 %.2f%%，因此被系统自动禁用。",
-		config.MetricQueueSize, successRate*100, config.MetricSuccessRateThreshold*100)
+	subject := fmt.Sprintf("渠道 #%d 已被禁用", channelId)
+	content := fmt.Sprintf("该渠道（#%d）在最近 %d 次调用中成功率为 %.2f%%，低于阈值 %.2f%%，因此被系统自动禁用。",
+		channelId, config.MetricQueueSize, successRate*100, config.MetricSuccessRateThreshold*100)
 	notifyRootUser(subject, content)
 }
 
@@ -49,7 +49,7 @@ func MetricDisableChannel(channelId int, successRate float64) {
 func EnableChannel(channelId int, channelName string) {
 	model.UpdateChannelStatusById(channelId, common.ChannelStatusEnabled)
 	logger.SysLog(fmt.Sprintf("channel #%d has been enabled", channelId))
-	subject := fmt.Sprintf("通道「%s」（#%d）已被启用", channelName, channelId)
-	content := fmt.Sprintf("通道「%s」（#%d）已被启用", channelName, channelId)
+	subject := fmt.Sprintf("渠道「%s」（#%d）已被启用", channelName, channelId)
+	content := fmt.Sprintf("渠道「%s」（#%d）已被启用", channelName, channelId)
 	notifyRootUser(subject, content)
 }

pull_request_template.md

@@ -1,9 +1,10 @@
 [//]: # (请按照以下格式关联 issue)
-[//]: # (请在提交 PR 前确认所提交的功能可用，附上截图即可，这将有助于项目维护者 review & merge 该 PR，谢谢)
+[//]: # (请在提交 PR 前确认所提交的功能可用，需要附上截图，谢谢)
 [//]: # (项目维护者一般仅在周末处理 PR，因此如若未能及时回复希望能理解)
 [//]: # (开发者交流群：910657413)
 [//]: # (请在提交 PR 之前删除上面的注释)
 
 close #issue_number
 
-我已确认该 PR 已自测通过，相关截图如下：
+我已确认该 PR 已自测通过，相关截图如下：
+（此处放上测试通过的截图，如果不涉及前端改动或从 UI 上无法看出，请放终端启动成功的截图）

relay/channel/aiproxy/adaptor.go

@@ -38,6 +38,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return aiProxyLibraryRequest, nil
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channel.DoRequestHelper(a, c, meta, requestBody)
 }

relay/channel/ali/adaptor.go

@@ -23,10 +23,16 @@ func (a *Adaptor) Init(meta *util.RelayMeta) {
 }
 
 func (a *Adaptor) GetRequestURL(meta *util.RelayMeta) (string, error) {
-	fullRequestURL := fmt.Sprintf("%s/api/v1/services/aigc/text-generation/generation", meta.BaseURL)
-	if meta.Mode == constant.RelayModeEmbeddings {
+	fullRequestURL := ""
+	switch meta.Mode {
+	case constant.RelayModeEmbeddings:
 		fullRequestURL = fmt.Sprintf("%s/api/v1/services/embeddings/text-embedding/text-embedding", meta.BaseURL)
+	case constant.RelayModeImagesGenerations:
+		fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/text2image/image-synthesis", meta.BaseURL)
+	default:
+		fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/text-generation/generation", meta.BaseURL)
 	}
+
 	return fullRequestURL, nil
 }
 
@@ -34,10 +40,12 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *ut
 	channel.SetupCommonRequestHeader(c, req, meta)
 	if meta.IsStream {
 		req.Header.Set("Accept", "text/event-stream")
+		req.Header.Set("X-DashScope-SSE", "enable")
 	}
 	req.Header.Set("Authorization", "Bearer "+meta.APIKey)
-	if meta.IsStream {
-		req.Header.Set("X-DashScope-SSE", "enable")
+
+	if meta.Mode == constant.RelayModeImagesGenerations {
+		req.Header.Set("X-DashScope-Async", "enable")
 	}
 	if c.GetString(common.ConfigKeyPlugin) != "" {
 		req.Header.Set("X-DashScope-Plugin", c.GetString(common.ConfigKeyPlugin))
@@ -51,14 +59,23 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	}
 	switch relayMode {
 	case constant.RelayModeEmbeddings:
-		baiduEmbeddingRequest := ConvertEmbeddingRequest(*request)
-		return baiduEmbeddingRequest, nil
+		aliEmbeddingRequest := ConvertEmbeddingRequest(*request)
+		return aliEmbeddingRequest, nil
 	default:
-		baiduRequest := ConvertRequest(*request)
-		return baiduRequest, nil
+		aliRequest := ConvertRequest(*request)
+		return aliRequest, nil
 	}
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+
+	aliRequest := ConvertImageRequest(*request)
+	return aliRequest, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channel.DoRequestHelper(a, c, meta, requestBody)
 }
@@ -70,6 +87,8 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *util.Rel
 		switch meta.Mode {
 		case constant.RelayModeEmbeddings:
 			err, usage = EmbeddingHandler(c, resp)
+		case constant.RelayModeImagesGenerations:
+			err, usage = ImageHandler(c, resp)
 		default:
 			err, usage = Handler(c, resp)
 		}

relay/channel/ali/constants.go

@@ -3,4 +3,5 @@ package ali
 var ModelList = []string{
 	"qwen-turbo", "qwen-plus", "qwen-max", "qwen-max-longcontext",
 	"text-embedding-v1",
+	"ali-stable-diffusion-xl", "ali-stable-diffusion-v1.5", "wanx-v1",
 }

relay/channel/ali/image.go

@@ -0,0 +1,192 @@
+package ali
+
+import (
+	"encoding/base64"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/common/helper"
+	"github.com/songquanpeng/one-api/common/logger"
+	"github.com/songquanpeng/one-api/relay/channel/openai"
+	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+	"strings"
+	"time"
+)
+
+func ImageHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	apiKey := c.Request.Header.Get("Authorization")
+	apiKey = strings.TrimPrefix(apiKey, "Bearer ")
+	responseFormat := c.GetString("response_format")
+
+	var aliTaskResponse TaskResponse
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = json.Unmarshal(responseBody, &aliTaskResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+
+	if aliTaskResponse.Message != "" {
+		logger.SysError("aliAsyncTask err: " + string(responseBody))
+		return openai.ErrorWrapper(errors.New(aliTaskResponse.Message), "ali_async_task_failed", http.StatusInternalServerError), nil
+	}
+
+	aliResponse, _, err := asyncTaskWait(aliTaskResponse.Output.TaskId, apiKey)
+	if err != nil {
+		return openai.ErrorWrapper(err, "ali_async_task_wait_failed", http.StatusInternalServerError), nil
+	}
+
+	if aliResponse.Output.TaskStatus != "SUCCEEDED" {
+		return &model.ErrorWithStatusCode{
+			Error: model.Error{
+				Message: aliResponse.Output.Message,
+				Type:    "ali_error",
+				Param:   "",
+				Code:    aliResponse.Output.Code,
+			},
+			StatusCode: resp.StatusCode,
+		}, nil
+	}
+
+	fullTextResponse := responseAli2OpenAIImage(aliResponse, responseFormat)
+	jsonResponse, err := json.Marshal(fullTextResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, err = c.Writer.Write(jsonResponse)
+	return nil, nil
+}
+
+func asyncTask(taskID string, key string) (*TaskResponse, error, []byte) {
+	url := fmt.Sprintf("https://dashscope.aliyuncs.com/api/v1/tasks/%s", taskID)
+
+	var aliResponse TaskResponse
+
+	req, err := http.NewRequest("GET", url, nil)
+	if err != nil {
+		return &aliResponse, err, nil
+	}
+
+	req.Header.Set("Authorization", "Bearer "+key)
+
+	client := &http.Client{}
+	resp, err := client.Do(req)
+	if err != nil {
+		logger.SysError("aliAsyncTask client.Do err: " + err.Error())
+		return &aliResponse, err, nil
+	}
+	defer resp.Body.Close()
+
+	responseBody, err := io.ReadAll(resp.Body)
+
+	var response TaskResponse
+	err = json.Unmarshal(responseBody, &response)
+	if err != nil {
+		logger.SysError("aliAsyncTask NewDecoder err: " + err.Error())
+		return &aliResponse, err, nil
+	}
+
+	return &response, nil, responseBody
+}
+
+func asyncTaskWait(taskID string, key string) (*TaskResponse, []byte, error) {
+	waitSeconds := 2
+	step := 0
+	maxStep := 20
+
+	var taskResponse TaskResponse
+	var responseBody []byte
+
+	for {
+		step++
+		rsp, err, body := asyncTask(taskID, key)
+		responseBody = body
+		if err != nil {
+			return &taskResponse, responseBody, err
+		}
+
+		if rsp.Output.TaskStatus == "" {
+			return &taskResponse, responseBody, nil
+		}
+
+		switch rsp.Output.TaskStatus {
+		case "FAILED":
+			fallthrough
+		case "CANCELED":
+			fallthrough
+		case "SUCCEEDED":
+			fallthrough
+		case "UNKNOWN":
+			return rsp, responseBody, nil
+		}
+		if step >= maxStep {
+			break
+		}
+		time.Sleep(time.Duration(waitSeconds) * time.Second)
+	}
+
+	return nil, nil, fmt.Errorf("aliAsyncTaskWait timeout")
+}
+
+func responseAli2OpenAIImage(response *TaskResponse, responseFormat string) *openai.ImageResponse {
+	imageResponse := openai.ImageResponse{
+		Created: helper.GetTimestamp(),
+	}
+
+	for _, data := range response.Output.Results {
+		var b64Json string
+		if responseFormat == "b64_json" {
+			// 读取 data.Url 的图片数据并转存到 b64Json
+			imageData, err := getImageData(data.Url)
+			if err != nil {
+				// 处理获取图片数据失败的情况
+				logger.SysError("getImageData Error getting image data: " + err.Error())
+				continue
+			}
+
+			// 将图片数据转为 Base64 编码的字符串
+			b64Json = Base64Encode(imageData)
+		} else {
+			// 如果 responseFormat 不是 "b64_json"，则直接使用 data.B64Image
+			b64Json = data.B64Image
+		}
+
+		imageResponse.Data = append(imageResponse.Data, openai.ImageData{
+			Url:           data.Url,
+			B64Json:       b64Json,
+			RevisedPrompt: "",
+		})
+	}
+	return &imageResponse
+}
+
+func getImageData(url string) ([]byte, error) {
+	response, err := http.Get(url)
+	if err != nil {
+		return nil, err
+	}
+	defer response.Body.Close()
+
+	imageData, err := io.ReadAll(response.Body)
+	if err != nil {
+		return nil, err
+	}
+
+	return imageData, nil
+}
+
+func Base64Encode(data []byte) string {
+	b64Json := base64.StdEncoding.EncodeToString(data)
+	return b64Json
+}

relay/channel/ali/main.go

@@ -48,6 +48,9 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
 			MaxTokens:         request.MaxTokens,
 			Temperature:       request.Temperature,
 			TopP:              request.TopP,
+			TopK:              request.TopK,
+			ResultFormat:      "message",
+			Tools:             request.Tools,
 		},
 	}
 }
@@ -63,6 +66,17 @@ func ConvertEmbeddingRequest(request model.GeneralOpenAIRequest) *EmbeddingReque
 	}
 }
 
+func ConvertImageRequest(request model.ImageRequest) *ImageRequest {
+	var imageRequest ImageRequest
+	imageRequest.Input.Prompt = request.Prompt
+	imageRequest.Model = request.Model
+	imageRequest.Parameters.Size = strings.Replace(request.Size, "x", "*", -1)
+	imageRequest.Parameters.N = request.N
+	imageRequest.ResponseFormat = request.ResponseFormat
+
+	return &imageRequest
+}
+
 func EmbeddingHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
 	var aliResponse EmbeddingResponse
 	err := json.NewDecoder(resp.Body).Decode(&aliResponse)
@@ -117,19 +131,11 @@ func embeddingResponseAli2OpenAI(response *EmbeddingResponse) *openai.EmbeddingR
 }
 
 func responseAli2OpenAI(response *ChatResponse) *openai.TextResponse {
-	choice := openai.TextResponseChoice{
-		Index: 0,
-		Message: model.Message{
-			Role:    "assistant",
-			Content: response.Output.Text,
-		},
-		FinishReason: response.Output.FinishReason,
-	}
 	fullTextResponse := openai.TextResponse{
 		Id:      response.RequestId,
 		Object:  "chat.completion",
 		Created: helper.GetTimestamp(),
-		Choices: []openai.TextResponseChoice{choice},
+		Choices: response.Output.Choices,
 		Usage: model.Usage{
 			PromptTokens:     response.Usage.InputTokens,
 			CompletionTokens: response.Usage.OutputTokens,
@@ -140,10 +146,14 @@ func responseAli2OpenAI(response *ChatResponse) *openai.TextResponse {
 }
 
 func streamResponseAli2OpenAI(aliResponse *ChatResponse) *openai.ChatCompletionsStreamResponse {
+	if len(aliResponse.Output.Choices) == 0 {
+		return nil
+	}
+	aliChoice := aliResponse.Output.Choices[0]
 	var choice openai.ChatCompletionsStreamResponseChoice
-	choice.Delta.Content = aliResponse.Output.Text
-	if aliResponse.Output.FinishReason != "null" {
-		finishReason := aliResponse.Output.FinishReason
+	choice.Delta = aliChoice.Message
+	if aliChoice.FinishReason != "null" {
+		finishReason := aliChoice.FinishReason
 		choice.FinishReason = &finishReason
 	}
 	response := openai.ChatCompletionsStreamResponse{
@@ -204,6 +214,9 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 				usage.TotalTokens = aliResponse.Usage.InputTokens + aliResponse.Usage.OutputTokens
 			}
 			response := streamResponseAli2OpenAI(&aliResponse)
+			if response == nil {
+				return true
+			}
 			//response.Choices[0].Delta.Content = strings.TrimPrefix(response.Choices[0].Delta.Content, lastResponseText)
 			//lastResponseText = aliResponse.Output.Text
 			jsonResponse, err := json.Marshal(response)
@@ -226,6 +239,7 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 }
 
 func Handler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	ctx := c.Request.Context()
 	var aliResponse ChatResponse
 	responseBody, err := io.ReadAll(resp.Body)
 	if err != nil {
@@ -235,6 +249,7 @@ func Handler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
 	}
+	logger.Debugf(ctx, "response body: %s\n", responseBody)
 	err = json.Unmarshal(responseBody, &aliResponse)
 	if err != nil {
 		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil

relay/channel/ali/model.go

@@ -1,5 +1,10 @@
 package ali
 
+import (
+	"github.com/songquanpeng/one-api/relay/channel/openai"
+	"github.com/songquanpeng/one-api/relay/model"
+)
+
 type Message struct {
 	Content string `json:"content"`
 	Role    string `json:"role"`
@@ -11,13 +16,15 @@ type Input struct {
 }
 
 type Parameters struct {
-	TopP              float64 `json:"top_p,omitempty"`
-	TopK              int     `json:"top_k,omitempty"`
-	Seed              uint64  `json:"seed,omitempty"`
-	EnableSearch      bool    `json:"enable_search,omitempty"`
-	IncrementalOutput bool    `json:"incremental_output,omitempty"`
-	MaxTokens         int     `json:"max_tokens,omitempty"`
-	Temperature       float64 `json:"temperature,omitempty"`
+	TopP              float64      `json:"top_p,omitempty"`
+	TopK              int          `json:"top_k,omitempty"`
+	Seed              uint64       `json:"seed,omitempty"`
+	EnableSearch      bool         `json:"enable_search,omitempty"`
+	IncrementalOutput bool         `json:"incremental_output,omitempty"`
+	MaxTokens         int          `json:"max_tokens,omitempty"`
+	Temperature       float64      `json:"temperature,omitempty"`
+	ResultFormat      string       `json:"result_format,omitempty"`
+	Tools             []model.Tool `json:"tools,omitempty"`
 }
 
 type ChatRequest struct {
@@ -26,6 +33,79 @@ type ChatRequest struct {
 	Parameters Parameters `json:"parameters,omitempty"`
 }
 
+type ImageRequest struct {
+	Model string `json:"model"`
+	Input struct {
+		Prompt         string `json:"prompt"`
+		NegativePrompt string `json:"negative_prompt,omitempty"`
+	} `json:"input"`
+	Parameters struct {
+		Size  string `json:"size,omitempty"`
+		N     int    `json:"n,omitempty"`
+		Steps string `json:"steps,omitempty"`
+		Scale string `json:"scale,omitempty"`
+	} `json:"parameters,omitempty"`
+	ResponseFormat string `json:"response_format,omitempty"`
+}
+
+type TaskResponse struct {
+	StatusCode int    `json:"status_code,omitempty"`
+	RequestId  string `json:"request_id,omitempty"`
+	Code       string `json:"code,omitempty"`
+	Message    string `json:"message,omitempty"`
+	Output     struct {
+		TaskId     string `json:"task_id,omitempty"`
+		TaskStatus string `json:"task_status,omitempty"`
+		Code       string `json:"code,omitempty"`
+		Message    string `json:"message,omitempty"`
+		Results    []struct {
+			B64Image string `json:"b64_image,omitempty"`
+			Url      string `json:"url,omitempty"`
+			Code     string `json:"code,omitempty"`
+			Message  string `json:"message,omitempty"`
+		} `json:"results,omitempty"`
+		TaskMetrics struct {
+			Total     int `json:"TOTAL,omitempty"`
+			Succeeded int `json:"SUCCEEDED,omitempty"`
+			Failed    int `json:"FAILED,omitempty"`
+		} `json:"task_metrics,omitempty"`
+	} `json:"output,omitempty"`
+	Usage Usage `json:"usage"`
+}
+
+type Header struct {
+	Action       string `json:"action,omitempty"`
+	Streaming    string `json:"streaming,omitempty"`
+	TaskID       string `json:"task_id,omitempty"`
+	Event        string `json:"event,omitempty"`
+	ErrorCode    string `json:"error_code,omitempty"`
+	ErrorMessage string `json:"error_message,omitempty"`
+	Attributes   any    `json:"attributes,omitempty"`
+}
+
+type Payload struct {
+	Model      string `json:"model,omitempty"`
+	Task       string `json:"task,omitempty"`
+	TaskGroup  string `json:"task_group,omitempty"`
+	Function   string `json:"function,omitempty"`
+	Parameters struct {
+		SampleRate int     `json:"sample_rate,omitempty"`
+		Rate       float64 `json:"rate,omitempty"`
+		Format     string  `json:"format,omitempty"`
+	} `json:"parameters,omitempty"`
+	Input struct {
+		Text string `json:"text,omitempty"`
+	} `json:"input,omitempty"`
+	Usage struct {
+		Characters int `json:"characters,omitempty"`
+	} `json:"usage,omitempty"`
+}
+
+type WSSMessage struct {
+	Header  Header  `json:"header,omitempty"`
+	Payload Payload `json:"payload,omitempty"`
+}
+
 type EmbeddingRequest struct {
 	Model string `json:"model"`
 	Input struct {
@@ -62,8 +142,9 @@ type Usage struct {
 }
 
 type Output struct {
-	Text         string `json:"text"`
-	FinishReason string `json:"finish_reason"`
+	//Text         string                      `json:"text"`
+	//FinishReason string                      `json:"finish_reason"`
+	Choices []openai.TextResponseChoice `json:"choices"`
 }
 
 type ChatResponse struct {

relay/channel/anthropic/adaptor.go

@@ -41,6 +41,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return ConvertRequest(*request), nil
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channel.DoRequestHelper(a, c, meta, requestBody)
 }

relay/channel/anthropic/main.go

@@ -38,6 +38,7 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request {
 		MaxTokens:   textRequest.MaxTokens,
 		Temperature: textRequest.Temperature,
 		TopP:        textRequest.TopP,
+		TopK:        textRequest.TopK,
 		Stream:      textRequest.Stream,
 	}
 	if claudeRequest.MaxTokens == 0 {

relay/channel/baidu/adaptor.go

@@ -38,16 +38,34 @@ func (a *Adaptor) GetRequestURL(meta *util.RelayMeta) (string, error) {
 		suffix += "completions_pro"
 	case "ERNIE-Bot-4":
 		suffix += "completions_pro"
-	case "ERNIE-3.5-8K":
-		suffix += "completions"
-	case "ERNIE-Bot-8K":
-		suffix += "ernie_bot_8k"
 	case "ERNIE-Bot":
 		suffix += "completions"
-	case "ERNIE-Speed":
-		suffix += "ernie_speed"
 	case "ERNIE-Bot-turbo":
 		suffix += "eb-instant"
+	case "ERNIE-Speed":
+		suffix += "ernie_speed"
+	case "ERNIE-4.0-8K":
+		suffix += "completions_pro"
+	case "ERNIE-3.5-8K":
+		suffix += "completions"
+	case "ERNIE-3.5-8K-0205":
+		suffix += "ernie-3.5-8k-0205"
+	case "ERNIE-3.5-8K-1222":
+		suffix += "ernie-3.5-8k-1222"
+	case "ERNIE-Bot-8K":
+		suffix += "ernie_bot_8k"
+	case "ERNIE-3.5-4K-0205":
+		suffix += "ernie-3.5-4k-0205"
+	case "ERNIE-Speed-8K":
+		suffix += "ernie_speed"
+	case "ERNIE-Speed-128K":
+		suffix += "ernie-speed-128k"
+	case "ERNIE-Lite-8K-0922":
+		suffix += "eb-instant"
+	case "ERNIE-Lite-8K-0308":
+		suffix += "ernie-lite-8k"
+	case "ERNIE-Tiny-8K":
+		suffix += "ernie-tiny-8k"
 	case "BLOOMZ-7B":
 		suffix += "bloomz_7b1"
 	case "Embedding-V1":
@@ -59,7 +77,7 @@ func (a *Adaptor) GetRequestURL(meta *util.RelayMeta) (string, error) {
 	case "tao-8k":
 		suffix += "tao_8k"
 	default:
-		suffix += meta.ActualModelName
+		suffix += strings.ToLower(meta.ActualModelName)
 	}
 	fullRequestURL := fmt.Sprintf("%s/rpc/2.0/ai_custom/v1/wenxinworkshop/%s", meta.BaseURL, suffix)
 	var accessToken string
@@ -91,6 +109,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	}
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channel.DoRequestHelper(a, c, meta, requestBody)
 }

relay/channel/baidu/constants.go

@@ -1,11 +1,18 @@
 package baidu
 
 var ModelList = []string{
-	"ERNIE-Bot-4",
+	"ERNIE-4.0-8K",
+	"ERNIE-3.5-8K",
+	"ERNIE-3.5-8K-0205",
+	"ERNIE-3.5-8K-1222",
 	"ERNIE-Bot-8K",
-	"ERNIE-Bot",
-	"ERNIE-Speed",
-	"ERNIE-Bot-turbo",
+	"ERNIE-3.5-4K-0205",
+	"ERNIE-Speed-8K",
+	"ERNIE-Speed-128K",
+	"ERNIE-Lite-8K-0922",
+	"ERNIE-Lite-8K-0308",
+	"ERNIE-Tiny-8K",
+	"BLOOMZ-7B",
 	"Embedding-V1",
 	"bge-large-zh",
 	"bge-large-en",

relay/channel/gemini/adaptor.go

@@ -42,6 +42,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return ConvertRequest(*request), nil
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channelhelper.DoRequestHelper(a, c, meta, requestBody)
 }

relay/channel/gemini/constants.go

@@ -1,6 +1,8 @@
 package gemini
 
+// https://ai.google.dev/models/gemini
+
 var ModelList = []string{
-	"gemini-pro", "gemini-1.0-pro-001",
+	"gemini-pro", "gemini-1.0-pro-001", "gemini-1.5-pro",
 	"gemini-pro-vision", "gemini-1.0-pro-vision-001",
 }

relay/channel/interface.go

@@ -13,6 +13,7 @@ type Adaptor interface {
 	GetRequestURL(meta *util.RelayMeta) (string, error)
 	SetupRequestHeader(c *gin.Context, req *http.Request, meta *util.RelayMeta) error
 	ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error)
+	ConvertImageRequest(request *model.ImageRequest) (any, error)
 	DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error)
 	DoResponse(c *gin.Context, resp *http.Response, meta *util.RelayMeta) (usage *model.Usage, err *model.ErrorWithStatusCode)
 	GetModelList() []string

relay/channel/ollama/adaptor.go

@@ -3,13 +3,14 @@ package ollama
 import (
 	"errors"
 	"fmt"
+	"io"
+	"net/http"
+
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/relay/channel"
 	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/util"
-	"io"
-	"net/http"
 )
 
 type Adaptor struct {
@@ -22,6 +23,9 @@ func (a *Adaptor) Init(meta *util.RelayMeta) {
 func (a *Adaptor) GetRequestURL(meta *util.RelayMeta) (string, error) {
 	// https://github.com/ollama/ollama/blob/main/docs/api.md
 	fullRequestURL := fmt.Sprintf("%s/api/chat", meta.BaseURL)
+	if meta.Mode == constant.RelayModeEmbeddings {
+		fullRequestURL = fmt.Sprintf("%s/api/embeddings", meta.BaseURL)
+	}
 	return fullRequestURL, nil
 }
 
@@ -37,12 +41,20 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	}
 	switch relayMode {
 	case constant.RelayModeEmbeddings:
-		return nil, errors.New("not supported")
+		ollamaEmbeddingRequest := ConvertEmbeddingRequest(*request)
+		return ollamaEmbeddingRequest, nil
 	default:
 		return ConvertRequest(*request), nil
 	}
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channel.DoRequestHelper(a, c, meta, requestBody)
 }
@@ -51,7 +63,12 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *util.Rel
 	if meta.IsStream {
 		err, usage = StreamHandler(c, resp)
 	} else {
-		err, usage = Handler(c, resp)
+		switch meta.Mode {
+		case constant.RelayModeEmbeddings:
+			err, usage = EmbeddingHandler(c, resp)
+		default:
+			err, usage = Handler(c, resp)
+		}
 	}
 	return
 }

relay/channel/ollama/main.go

@@ -5,6 +5,10 @@ import (
 	"context"
 	"encoding/json"
 	"fmt"
+	"io"
+	"net/http"
+	"strings"
+
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/helper"
@@ -12,9 +16,6 @@ import (
 	"github.com/songquanpeng/one-api/relay/channel/openai"
 	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
-	"io"
-	"net/http"
-	"strings"
 )
 
 func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
@@ -139,6 +140,64 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 	return nil, &usage
 }
 
+func ConvertEmbeddingRequest(request model.GeneralOpenAIRequest) *EmbeddingRequest {
+	return &EmbeddingRequest{
+		Model:  request.Model,
+		Prompt: strings.Join(request.ParseInput(), " "),
+	}
+}
+
+func EmbeddingHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	var ollamaResponse EmbeddingResponse
+	err := json.NewDecoder(resp.Body).Decode(&ollamaResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+
+	if ollamaResponse.Error != "" {
+		return &model.ErrorWithStatusCode{
+			Error: model.Error{
+				Message: ollamaResponse.Error,
+				Type:    "ollama_error",
+				Param:   "",
+				Code:    "ollama_error",
+			},
+			StatusCode: resp.StatusCode,
+		}, nil
+	}
+
+	fullTextResponse := embeddingResponseOllama2OpenAI(&ollamaResponse)
+	jsonResponse, err := json.Marshal(fullTextResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, err = c.Writer.Write(jsonResponse)
+	return nil, &fullTextResponse.Usage
+}
+
+func embeddingResponseOllama2OpenAI(response *EmbeddingResponse) *openai.EmbeddingResponse {
+	openAIEmbeddingResponse := openai.EmbeddingResponse{
+		Object: "list",
+		Data:   make([]openai.EmbeddingResponseItem, 0, 1),
+		Model:  "text-embedding-v1",
+		Usage:  model.Usage{TotalTokens: 0},
+	}
+
+	openAIEmbeddingResponse.Data = append(openAIEmbeddingResponse.Data, openai.EmbeddingResponseItem{
+		Object:    `embedding`,
+		Index:     0,
+		Embedding: response.Embedding,
+	})
+	return &openAIEmbeddingResponse
+}
+
 func Handler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
 	ctx := context.TODO()
 	var ollamaResponse ChatResponse

relay/channel/ollama/model.go

@@ -35,3 +35,13 @@ type ChatResponse struct {
 	EvalDuration    int     `json:"eval_duration,omitempty"`
 	Error           string  `json:"error,omitempty"`
 }
+
+type EmbeddingRequest struct {
+	Model  string `json:"model"`
+	Prompt string `json:"prompt"`
+}
+
+type EmbeddingResponse struct {
+	Error     string    `json:"error,omitempty"`
+	Embedding []float64 `json:"embedding,omitempty"`
+}

relay/channel/openai/adaptor.go

@@ -7,6 +7,7 @@ import (
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/relay/channel"
 	"github.com/songquanpeng/one-api/relay/channel/minimax"
+	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/util"
 	"io"
@@ -25,17 +26,21 @@ func (a *Adaptor) Init(meta *util.RelayMeta) {
 func (a *Adaptor) GetRequestURL(meta *util.RelayMeta) (string, error) {
 	switch meta.ChannelType {
 	case common.ChannelTypeAzure:
+		if meta.Mode == constant.RelayModeImagesGenerations {
+			// https://learn.microsoft.com/en-us/azure/ai-services/openai/dall-e-quickstart?tabs=dalle3%2Ccommand-line&pivots=rest-api
+			// https://{resource_name}.openai.azure.com/openai/deployments/dall-e-3/images/generations?api-version=2024-03-01-preview
+			fullRequestURL := fmt.Sprintf("%s/openai/deployments/%s/images/generations?api-version=%s", meta.BaseURL, meta.ActualModelName, meta.APIVersion)
+			return fullRequestURL, nil
+		}
+
 		// https://learn.microsoft.com/en-us/azure/cognitive-services/openai/chatgpt-quickstart?pivots=rest-api&tabs=command-line#rest-api
 		requestURL := strings.Split(meta.RequestURLPath, "?")[0]
 		requestURL = fmt.Sprintf("%s?api-version=%s", requestURL, meta.APIVersion)
 		task := strings.TrimPrefix(requestURL, "/v1/")
 		model_ := meta.ActualModelName
 		model_ = strings.Replace(model_, ".", "", -1)
-		// https://github.com/songquanpeng/one-api/issues/67
-		model_ = strings.TrimSuffix(model_, "-0301")
-		model_ = strings.TrimSuffix(model_, "-0314")
-		model_ = strings.TrimSuffix(model_, "-0613")
-
+		//https://github.com/songquanpeng/one-api/issues/1191
+		// {your endpoint}/openai/deployments/{your azure_model}/chat/completions?api-version={api_version}
 		requestURL = fmt.Sprintf("/openai/deployments/%s/%s", model_, task)
 		return util.GetFullRequestURL(meta.BaseURL, requestURL, meta.ChannelType), nil
 	case common.ChannelTypeMinimax:
@@ -66,6 +71,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return request, nil
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channel.DoRequestHelper(a, c, meta, requestBody)
 }
@@ -73,10 +85,17 @@ func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io
 func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *util.RelayMeta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
 	if meta.IsStream {
 		var responseText string
-		err, responseText, _ = StreamHandler(c, resp, meta.Mode)
-		usage = ResponseText2Usage(responseText, meta.ActualModelName, meta.PromptTokens)
+		err, responseText, usage = StreamHandler(c, resp, meta.Mode)
+		if usage == nil {
+			usage = ResponseText2Usage(responseText, meta.ActualModelName, meta.PromptTokens)
+		}
 	} else {
-		err, usage = Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
+		switch meta.Mode {
+		case constant.RelayModeImagesGenerations:
+			err, _ = ImageHandler(c, resp)
+		default:
+			err, usage = Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
+		}
 	}
 	return
 }

relay/channel/openai/compatible.go

@@ -9,6 +9,7 @@ import (
 	"github.com/songquanpeng/one-api/relay/channel/minimax"
 	"github.com/songquanpeng/one-api/relay/channel/mistral"
 	"github.com/songquanpeng/one-api/relay/channel/moonshot"
+	"github.com/songquanpeng/one-api/relay/channel/stepfun"
 )
 
 var CompatibleChannels = []int{
@@ -20,6 +21,7 @@ var CompatibleChannels = []int{
 	common.ChannelTypeMistral,
 	common.ChannelTypeGroq,
 	common.ChannelTypeLingYiWanWu,
+	common.ChannelTypeStepFun,
 }
 
 func GetCompatibleChannelMeta(channelType int) (string, []string) {
@@ -40,6 +42,8 @@ func GetCompatibleChannelMeta(channelType int) (string, []string) {
 		return "groq", groq.ModelList
 	case common.ChannelTypeLingYiWanWu:
 		return "lingyiwanwu", lingyiwanwu.ModelList
+	case common.ChannelTypeStepFun:
+		return "stepfun", stepfun.ModelList
 	default:
 		return "openai", ModelList
 	}

relay/channel/openai/main.go

@@ -6,6 +6,7 @@ import (
 	"encoding/json"
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
+	"github.com/songquanpeng/one-api/common/conv"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
@@ -53,7 +54,7 @@ func StreamHandler(c *gin.Context, resp *http.Response, relayMode int) (*model.E
 						continue // just ignore the error
 					}
 					for _, choice := range streamResponse.Choices {
-						responseText += choice.Delta.Content
+						responseText += conv.AsString(choice.Delta.Content)
 					}
 					if streamResponse.Usage != nil {
 						usage = streamResponse.Usage
@@ -148,3 +149,37 @@ func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName st
 	}
 	return nil, &textResponse.Usage
 }
+
+func ImageHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	var imageResponse ImageResponse
+	responseBody, err := io.ReadAll(resp.Body)
+
+	if err != nil {
+		return ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = json.Unmarshal(responseBody, &imageResponse)
+	if err != nil {
+		return ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+
+	resp.Body = io.NopCloser(bytes.NewBuffer(responseBody))
+
+	for k, v := range resp.Header {
+		c.Writer.Header().Set(k, v[0])
+	}
+	c.Writer.WriteHeader(resp.StatusCode)
+
+	_, err = io.Copy(c.Writer, resp.Body)
+	if err != nil {
+		return ErrorWrapper(err, "copy_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	return nil, nil
+}

relay/channel/openai/model.go

@@ -110,20 +110,22 @@ type EmbeddingResponse struct {
 	model.Usage `json:"usage"`
 }
 
+type ImageData struct {
+	Url           string `json:"url,omitempty"`
+	B64Json       string `json:"b64_json,omitempty"`
+	RevisedPrompt string `json:"revised_prompt,omitempty"`
+}
+
 type ImageResponse struct {
-	Created int `json:"created"`
-	Data    []struct {
-		Url string `json:"url"`
-	}
+	Created int64       `json:"created"`
+	Data    []ImageData `json:"data"`
+	//model.Usage `json:"usage"`
 }
 
 type ChatCompletionsStreamResponseChoice struct {
-	Index int `json:"index"`
-	Delta struct {
-		Content string `json:"content"`
-		Role    string `json:"role,omitempty"`
-	} `json:"delta"`
-	FinishReason *string `json:"finish_reason,omitempty"`
+	Index        int           `json:"index"`
+	Delta        model.Message `json:"delta"`
+	FinishReason *string       `json:"finish_reason,omitempty"`
 }
 
 type ChatCompletionsStreamResponse struct {

relay/channel/palm/adaptor.go

@@ -36,6 +36,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return ConvertRequest(*request), nil
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channel.DoRequestHelper(a, c, meta, requestBody)
 }

relay/channel/stepfun/constants.go

@@ -0,0 +1,7 @@
+package stepfun
+
+var ModelList = []string{
+	"step-1-32k",
+	"step-1v-32k",
+	"step-1-200k",
+}

relay/channel/tencent/adaptor.go

@@ -52,6 +52,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return tencentRequest, nil
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channel.DoRequestHelper(a, c, meta, requestBody)
 }

relay/channel/tencent/main.go

@@ -10,6 +10,7 @@ import (
 	"fmt"
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
+	"github.com/songquanpeng/one-api/common/conv"
 	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/relay/channel/openai"
@@ -129,7 +130,7 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 			}
 			response := streamResponseTencent2OpenAI(&TencentResponse)
 			if len(response.Choices) != 0 {
-				responseText += response.Choices[0].Delta.Content
+				responseText += conv.AsString(response.Choices[0].Delta.Content)
 			}
 			jsonResponse, err := json.Marshal(response)
 			if err != nil {

relay/channel/xunfei/adaptor.go

@@ -38,6 +38,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	// xunfei's request is not http request, so we don't need to do anything here
 	dummyResp := &http.Response{}

relay/channel/xunfei/main.go

@@ -26,7 +26,11 @@ import (
 
 func requestOpenAI2Xunfei(request model.GeneralOpenAIRequest, xunfeiAppId string, domain string) *ChatRequest {
 	messages := make([]Message, 0, len(request.Messages))
+	var lastToolCalls []model.Tool
 	for _, message := range request.Messages {
+		if message.ToolCalls != nil {
+			lastToolCalls = message.ToolCalls
+		}
 		messages = append(messages, Message{
 			Role:    message.Role,
 			Content: message.StringContent(),
@@ -39,9 +43,33 @@ func requestOpenAI2Xunfei(request model.GeneralOpenAIRequest, xunfeiAppId string
 	xunfeiRequest.Parameter.Chat.TopK = request.N
 	xunfeiRequest.Parameter.Chat.MaxTokens = request.MaxTokens
 	xunfeiRequest.Payload.Message.Text = messages
+	if len(lastToolCalls) != 0 {
+		for _, toolCall := range lastToolCalls {
+			xunfeiRequest.Payload.Functions.Text = append(xunfeiRequest.Payload.Functions.Text, toolCall.Function)
+		}
+	}
+
 	return &xunfeiRequest
 }
 
+func getToolCalls(response *ChatResponse) []model.Tool {
+	var toolCalls []model.Tool
+	if len(response.Payload.Choices.Text) == 0 {
+		return toolCalls
+	}
+	item := response.Payload.Choices.Text[0]
+	if item.FunctionCall == nil {
+		return toolCalls
+	}
+	toolCall := model.Tool{
+		Id:       fmt.Sprintf("call_%s", helper.GetUUID()),
+		Type:     "function",
+		Function: *item.FunctionCall,
+	}
+	toolCalls = append(toolCalls, toolCall)
+	return toolCalls
+}
+
 func responseXunfei2OpenAI(response *ChatResponse) *openai.TextResponse {
 	if len(response.Payload.Choices.Text) == 0 {
 		response.Payload.Choices.Text = []ChatResponseTextItem{
@@ -53,8 +81,9 @@ func responseXunfei2OpenAI(response *ChatResponse) *openai.TextResponse {
 	choice := openai.TextResponseChoice{
 		Index: 0,
 		Message: model.Message{
-			Role:    "assistant",
-			Content: response.Payload.Choices.Text[0].Content,
+			Role:      "assistant",
+			Content:   response.Payload.Choices.Text[0].Content,
+			ToolCalls: getToolCalls(response),
 		},
 		FinishReason: constant.StopFinishReason,
 	}
@@ -78,6 +107,7 @@ func streamResponseXunfei2OpenAI(xunfeiResponse *ChatResponse) *openai.ChatCompl
 	}
 	var choice openai.ChatCompletionsStreamResponseChoice
 	choice.Delta.Content = xunfeiResponse.Payload.Choices.Text[0].Content
+	choice.Delta.ToolCalls = getToolCalls(xunfeiResponse)
 	if xunfeiResponse.Payload.Choices.Status == 2 {
 		choice.FinishReason = &constant.StopFinishReason
 	}
@@ -121,7 +151,7 @@ func StreamHandler(c *gin.Context, textRequest model.GeneralOpenAIRequest, appId
 	domain, authUrl := getXunfeiAuthUrl(c, apiKey, apiSecret, textRequest.Model)
 	dataChan, stopChan, err := xunfeiMakeRequest(textRequest, domain, authUrl, appId)
 	if err != nil {
-		return openai.ErrorWrapper(err, "make xunfei request err", http.StatusInternalServerError), nil
+		return openai.ErrorWrapper(err, "xunfei_request_failed", http.StatusInternalServerError), nil
 	}
 	common.SetEventStreamHeaders(c)
 	var usage model.Usage
@@ -151,7 +181,7 @@ func Handler(c *gin.Context, textRequest model.GeneralOpenAIRequest, appId strin
 	domain, authUrl := getXunfeiAuthUrl(c, apiKey, apiSecret, textRequest.Model)
 	dataChan, stopChan, err := xunfeiMakeRequest(textRequest, domain, authUrl, appId)
 	if err != nil {
-		return openai.ErrorWrapper(err, "make xunfei request err", http.StatusInternalServerError), nil
+		return openai.ErrorWrapper(err, "xunfei_request_failed", http.StatusInternalServerError), nil
 	}
 	var usage model.Usage
 	var content string
@@ -171,11 +201,7 @@ func Handler(c *gin.Context, textRequest model.GeneralOpenAIRequest, appId strin
 		}
 	}
 	if len(xunfeiResponse.Payload.Choices.Text) == 0 {
-		xunfeiResponse.Payload.Choices.Text = []ChatResponseTextItem{
-			{
-				Content: "",
-			},
-		}
+		return openai.ErrorWrapper(err, "xunfei_empty_response_detected", http.StatusInternalServerError), nil
 	}
 	xunfeiResponse.Payload.Choices.Text[0].Content = content
 
@@ -202,15 +228,21 @@ func xunfeiMakeRequest(textRequest model.GeneralOpenAIRequest, domain, authUrl,
 	if err != nil {
 		return nil, nil, err
 	}
+	_, msg, err := conn.ReadMessage()
+	if err != nil {
+		return nil, nil, err
+	}
 
 	dataChan := make(chan ChatResponse)
 	stopChan := make(chan bool)
 	go func() {
 		for {
-			_, msg, err := conn.ReadMessage()
-			if err != nil {
-				logger.SysError("error reading stream response: " + err.Error())
-				break
+			if msg == nil {
+				_, msg, err = conn.ReadMessage()
+				if err != nil {
+					logger.SysError("error reading stream response: " + err.Error())
+					break
+				}
 			}
 			var response ChatResponse
 			err = json.Unmarshal(msg, &response)
@@ -218,6 +250,7 @@ func xunfeiMakeRequest(textRequest model.GeneralOpenAIRequest, domain, authUrl,
 				logger.SysError("error unmarshalling stream response: " + err.Error())
 				break
 			}
+			msg = nil
 			dataChan <- response
 			if response.Payload.Choices.Status == 2 {
 				err := conn.Close()

relay/channel/xunfei/model.go

@@ -26,13 +26,18 @@ type ChatRequest struct {
 		Message struct {
 			Text []Message `json:"text"`
 		} `json:"message"`
+		Functions struct {
+			Text []model.Function `json:"text,omitempty"`
+		} `json:"functions,omitempty"`
 	} `json:"payload"`
 }
 
 type ChatResponseTextItem struct {
-	Content string `json:"content"`
-	Role    string `json:"role"`
-	Index   int    `json:"index"`
+	Content      string          `json:"content"`
+	Role         string          `json:"role"`
+	Index        int             `json:"index"`
+	ContentType  string          `json:"content_type"`
+	FunctionCall *model.Function `json:"function_call"`
 }
 
 type ChatResponse struct {

relay/channel/zhipu/adaptor.go

@@ -6,9 +6,11 @@ import (
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/relay/channel"
 	"github.com/songquanpeng/one-api/relay/channel/openai"
+	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/util"
 	"io"
+	"math"
 	"net/http"
 	"strings"
 )
@@ -34,6 +36,9 @@ func (a *Adaptor) GetRequestURL(meta *util.RelayMeta) (string, error) {
 	if a.APIVersion == "v4" {
 		return fmt.Sprintf("%s/api/paas/v4/chat/completions", meta.BaseURL), nil
 	}
+	if meta.Mode == constant.RelayModeEmbeddings {
+		return fmt.Sprintf("%s/api/paas/v4/embeddings", meta.BaseURL), nil
+	}
 	method := "invoke"
 	if meta.IsStream {
 		method = "sse-invoke"
@@ -52,14 +57,31 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	if request == nil {
 		return nil, errors.New("request is nil")
 	}
-	if request.TopP >= 1 {
-		request.TopP = 0.99
+	switch relayMode {
+	case constant.RelayModeEmbeddings:
+		baiduEmbeddingRequest := ConvertEmbeddingRequest(*request)
+		return baiduEmbeddingRequest, nil
+	default:
+		// TopP (0.0, 1.0)
+		request.TopP = math.Min(0.99, request.TopP)
+		request.TopP = math.Max(0.01, request.TopP)
+
+		// Temperature (0.0, 1.0)
+		request.Temperature = math.Min(0.99, request.Temperature)
+		request.Temperature = math.Max(0.01, request.Temperature)
+		a.SetVersionByModeName(request.Model)
+		if a.APIVersion == "v4" {
+			return request, nil
+		}
+		return ConvertRequest(*request), nil
 	}
-	a.SetVersionByModeName(request.Model)
-	if a.APIVersion == "v4" {
-		return request, nil
+}
+
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
 	}
-	return ConvertRequest(*request), nil
+	return request, nil
 }
 
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
@@ -79,14 +101,26 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *util.Rel
 	if a.APIVersion == "v4" {
 		return a.DoResponseV4(c, resp, meta)
 	}
+
 	if meta.IsStream {
 		err, usage = StreamHandler(c, resp)
 	} else {
-		err, usage = Handler(c, resp)
+		if meta.Mode == constant.RelayModeEmbeddings {
+			err, usage = EmbeddingsHandler(c, resp)
+		} else {
+			err, usage = Handler(c, resp)
+		}
 	}
 	return
 }
 
+func ConvertEmbeddingRequest(request model.GeneralOpenAIRequest) *EmbeddingRequest {
+	return &EmbeddingRequest{
+		Model: "embedding-2",
+		Input: request.Input.(string),
+	}
+}
+
 func (a *Adaptor) GetModelList() []string {
 	return ModelList
 }

relay/channel/zhipu/constants.go

@@ -2,5 +2,5 @@ package zhipu
 
 var ModelList = []string{
 	"chatglm_turbo", "chatglm_pro", "chatglm_std", "chatglm_lite",
-	"glm-4", "glm-4v", "glm-3-turbo",
+	"glm-4", "glm-4v", "glm-3-turbo", "embedding-2",
 }

relay/channel/zhipu/main.go

@@ -254,3 +254,50 @@ func Handler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *
 	_, err = c.Writer.Write(jsonResponse)
 	return nil, &fullTextResponse.Usage
 }
+
+func EmbeddingsHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	var zhipuResponse EmbeddingRespone
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = json.Unmarshal(responseBody, &zhipuResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	fullTextResponse := embeddingResponseZhipu2OpenAI(&zhipuResponse)
+	jsonResponse, err := json.Marshal(fullTextResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, err = c.Writer.Write(jsonResponse)
+	return nil, &fullTextResponse.Usage
+}
+
+func embeddingResponseZhipu2OpenAI(response *EmbeddingRespone) *openai.EmbeddingResponse {
+	openAIEmbeddingResponse := openai.EmbeddingResponse{
+		Object: "list",
+		Data:   make([]openai.EmbeddingResponseItem, 0, len(response.Embeddings)),
+		Model:  response.Model,
+		Usage: model.Usage{
+			PromptTokens:     response.PromptTokens,
+			CompletionTokens: response.CompletionTokens,
+			TotalTokens:      response.Usage.TotalTokens,
+		},
+	}
+
+	for _, item := range response.Embeddings {
+		openAIEmbeddingResponse.Data = append(openAIEmbeddingResponse.Data, openai.EmbeddingResponseItem{
+			Object:    `embedding`,
+			Index:     item.Index,
+			Embedding: item.Embedding,
+		})
+	}
+	return &openAIEmbeddingResponse
+}

relay/channel/zhipu/model.go

@@ -44,3 +44,21 @@ type tokenData struct {
 	Token      string
 	ExpiryTime time.Time
 }
+
+type EmbeddingRequest struct {
+	Model string `json:"model"`
+	Input string `json:"input"`
+}
+
+type EmbeddingRespone struct {
+	Model       string          `json:"model"`
+	Object      string          `json:"object"`
+	Embeddings  []EmbeddingData `json:"data"`
+	model.Usage `json:"usage"`
+}
+
+type EmbeddingData struct {
+	Index     int       `json:"index"`
+	Object    string    `json:"object"`
+	Embedding []float64 `json:"embedding"`
+}

relay/constant/image.go

@@ -11,14 +11,31 @@ var DalleSizeRatios = map[string]map[string]float64{
 		"1024x1792": 2,
 		"1792x1024": 2,
 	},
+	"stable-diffusion-xl": {
+		"512x1024":  1,
+		"1024x768":  1,
+		"1024x1024": 1,
+		"576x1024":  1,
+		"1024x576":  1,
+	},
+	"wanx-v1": {
+		"1024x1024": 1,
+		"720x1280":  1,
+		"1280x720":  1,
+	},
 }
 
 var DalleGenerationImageAmounts = map[string][2]int{
-	"dall-e-2": {1, 10},
-	"dall-e-3": {1, 1}, // OpenAI allows n=1 currently.
+	"dall-e-2":            {1, 10},
+	"dall-e-3":            {1, 1}, // OpenAI allows n=1 currently.
+	"stable-diffusion-xl": {1, 4}, // Ali
+	"wanx-v1":             {1, 4}, // Ali
 }
 
 var DalleImagePromptLengthLimitations = map[string]int{
-	"dall-e-2": 1000,
-	"dall-e-3": 4000,
+	"dall-e-2":            1000,
+	"dall-e-3":            4000,
+	"stable-diffusion-xl": 4000,
+	"wanx-v1":             4000,
+	"cogview-3":           833,
 }

relay/controller/audio.go

@@ -83,6 +83,24 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 			return openai.ErrorWrapper(err, "pre_consume_token_quota_failed", http.StatusForbidden)
 		}
 	}
+	succeed := false
+	defer func() {
+		if succeed {
+			return
+		}
+		if preConsumedQuota > 0 {
+			// we need to roll back the pre-consumed quota
+			defer func(ctx context.Context) {
+				go func() {
+					// negative means add quota back for token & user
+					err := model.PostConsumeTokenQuota(tokenId, -preConsumedQuota)
+					if err != nil {
+						logger.Error(ctx, fmt.Sprintf("error rollback pre-consumed quota: %s", err.Error()))
+					}
+				}()
+			}(c.Request.Context())
+		}
+	}()
 
 	// map model name
 	modelMapping := c.GetString("model_mapping")
@@ -104,10 +122,15 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 	}
 
 	fullRequestURL := util.GetFullRequestURL(baseURL, requestURL, channelType)
-	if relayMode == constant.RelayModeAudioTranscription && channelType == common.ChannelTypeAzure {
-		// https://learn.microsoft.com/en-us/azure/ai-services/openai/whisper-quickstart?tabs=command-line#rest-api
+	if channelType == common.ChannelTypeAzure {
 		apiVersion := util.GetAzureAPIVersion(c)
-		fullRequestURL = fmt.Sprintf("%s/openai/deployments/%s/audio/transcriptions?api-version=%s", baseURL, audioModel, apiVersion)
+		if relayMode == constant.RelayModeAudioTranscription {
+			// https://learn.microsoft.com/en-us/azure/ai-services/openai/whisper-quickstart?tabs=command-line#rest-api
+			fullRequestURL = fmt.Sprintf("%s/openai/deployments/%s/audio/transcriptions?api-version=%s", baseURL, audioModel, apiVersion)
+		} else if relayMode == constant.RelayModeAudioSpeech {
+			// https://learn.microsoft.com/en-us/azure/ai-services/openai/text-to-speech-quickstart?tabs=command-line#rest-api
+			fullRequestURL = fmt.Sprintf("%s/openai/deployments/%s/audio/speech?api-version=%s", baseURL, audioModel, apiVersion)
+		}
 	}
 
 	requestBody := &bytes.Buffer{}
@@ -123,7 +146,7 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		return openai.ErrorWrapper(err, "new_request_failed", http.StatusInternalServerError)
 	}
 
-	if relayMode == constant.RelayModeAudioTranscription && channelType == common.ChannelTypeAzure {
+	if (relayMode == constant.RelayModeAudioTranscription || relayMode == constant.RelayModeAudioSpeech) && channelType == common.ChannelTypeAzure {
 		// https://learn.microsoft.com/en-us/azure/ai-services/openai/whisper-quickstart?tabs=command-line#rest-api
 		apiKey := c.Request.Header.Get("Authorization")
 		apiKey = strings.TrimPrefix(apiKey, "Bearer ")
@@ -188,20 +211,9 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		resp.Body = io.NopCloser(bytes.NewBuffer(responseBody))
 	}
 	if resp.StatusCode != http.StatusOK {
-		if preConsumedQuota > 0 {
-			// we need to roll back the pre-consumed quota
-			defer func(ctx context.Context) {
-				go func() {
-					// negative means add quota back for token & user
-					err := model.PostConsumeTokenQuota(tokenId, -preConsumedQuota)
-					if err != nil {
-						logger.Error(ctx, fmt.Sprintf("error rollback pre-consumed quota: %s", err.Error()))
-					}
-				}()
-			}(c.Request.Context())
-		}
 		return util.RelayErrorHandler(resp)
 	}
+	succeed = true
 	quotaDelta := quota - preConsumedQuota
 	defer func(ctx context.Context) {
 		go util.PostConsumeQuota(ctx, tokenId, quotaDelta, quota, userId, channelId, modelRatio, groupRatio, audioModel, tokenName)

relay/controller/helper.go

@@ -36,8 +36,8 @@ func getAndValidateTextRequest(c *gin.Context, relayMode int) (*relaymodel.Gener
 	return textRequest, nil
 }
 
-func getImageRequest(c *gin.Context, relayMode int) (*openai.ImageRequest, error) {
-	imageRequest := &openai.ImageRequest{}
+func getImageRequest(c *gin.Context, relayMode int) (*relaymodel.ImageRequest, error) {
+	imageRequest := &relaymodel.ImageRequest{}
 	err := common.UnmarshalBodyReusable(c, imageRequest)
 	if err != nil {
 		return nil, err
@@ -54,7 +54,7 @@ func getImageRequest(c *gin.Context, relayMode int) (*openai.ImageRequest, error
 	return imageRequest, nil
 }
 
-func validateImageRequest(imageRequest *openai.ImageRequest, meta *util.RelayMeta) *relaymodel.ErrorWithStatusCode {
+func validateImageRequest(imageRequest *relaymodel.ImageRequest, meta *util.RelayMeta) *relaymodel.ErrorWithStatusCode {
 	// model validation
 	_, hasValidSize := constant.DalleSizeRatios[imageRequest.Model][imageRequest.Size]
 	if !hasValidSize {
@@ -77,7 +77,7 @@ func validateImageRequest(imageRequest *openai.ImageRequest, meta *util.RelayMet
 	return nil
 }
 
-func getImageCostRatio(imageRequest *openai.ImageRequest) (float64, error) {
+func getImageCostRatio(imageRequest *relaymodel.ImageRequest) (float64, error) {
 	if imageRequest == nil {
 		return 0, errors.New("imageRequest is nil")
 	}

relay/controller/image.go

@@ -6,18 +6,17 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/model"
 	"github.com/songquanpeng/one-api/relay/channel/openai"
 	"github.com/songquanpeng/one-api/relay/constant"
+	"github.com/songquanpeng/one-api/relay/helper"
 	relaymodel "github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/util"
 	"io"
 	"net/http"
-	"strings"
-
-	"github.com/gin-gonic/gin"
 )
 
 func isWithinRange(element string, value int) bool {
@@ -56,15 +55,6 @@ func RelayImageHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		return openai.ErrorWrapper(err, "get_image_cost_ratio_failed", http.StatusInternalServerError)
 	}
 
-	requestURL := c.Request.URL.String()
-	fullRequestURL := util.GetFullRequestURL(meta.BaseURL, requestURL, meta.ChannelType)
-	if meta.ChannelType == common.ChannelTypeAzure {
-		// https://learn.microsoft.com/en-us/azure/ai-services/openai/dall-e-quickstart?tabs=dalle3%2Ccommand-line&pivots=rest-api
-		apiVersion := util.GetAzureAPIVersion(c)
-		// https://{resource_name}.openai.azure.com/openai/deployments/dall-e-3/images/generations?api-version=2023-06-01-preview
-		fullRequestURL = fmt.Sprintf("%s/openai/deployments/%s/images/generations?api-version=%s", meta.BaseURL, imageRequest.Model, apiVersion)
-	}
-
 	var requestBody io.Reader
 	if isModelMapped || meta.ChannelType == common.ChannelTypeAzure { // make Azure channel request body
 		jsonStr, err := json.Marshal(imageRequest)
@@ -76,6 +66,29 @@ func RelayImageHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		requestBody = c.Request.Body
 	}
 
+	adaptor := helper.GetAdaptor(meta.APIType)
+	if adaptor == nil {
+		return openai.ErrorWrapper(fmt.Errorf("invalid api type: %d", meta.APIType), "invalid_api_type", http.StatusBadRequest)
+	}
+
+	switch meta.ChannelType {
+	case common.ChannelTypeAli:
+		fallthrough
+	case common.ChannelTypeBaidu:
+		fallthrough
+	case common.ChannelTypeZhipu:
+		finalRequest, err := adaptor.ConvertImageRequest(imageRequest)
+		if err != nil {
+			return openai.ErrorWrapper(err, "convert_image_request_failed", http.StatusInternalServerError)
+		}
+
+		jsonStr, err := json.Marshal(finalRequest)
+		if err != nil {
+			return openai.ErrorWrapper(err, "marshal_image_request_failed", http.StatusInternalServerError)
+		}
+		requestBody = bytes.NewBuffer(jsonStr)
+	}
+
 	modelRatio := common.GetModelRatio(imageRequest.Model)
 	groupRatio := common.GetGroupRatio(meta.Group)
 	ratio := modelRatio * groupRatio
@@ -87,36 +100,13 @@ func RelayImageHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		return openai.ErrorWrapper(errors.New("user quota is not enough"), "insufficient_user_quota", http.StatusForbidden)
 	}
 
-	req, err := http.NewRequest(c.Request.Method, fullRequestURL, requestBody)
-	if err != nil {
-		return openai.ErrorWrapper(err, "new_request_failed", http.StatusInternalServerError)
-	}
-	token := c.Request.Header.Get("Authorization")
-	if meta.ChannelType == common.ChannelTypeAzure { // Azure authentication
-		token = strings.TrimPrefix(token, "Bearer ")
-		req.Header.Set("api-key", token)
-	} else {
-		req.Header.Set("Authorization", token)
-	}
-
-	req.Header.Set("Content-Type", c.Request.Header.Get("Content-Type"))
-	req.Header.Set("Accept", c.Request.Header.Get("Accept"))
-
-	resp, err := util.HTTPClient.Do(req)
+	// do request
+	resp, err := adaptor.DoRequest(c, meta, requestBody)
 	if err != nil {
+		logger.Errorf(ctx, "DoRequest failed: %s", err.Error())
 		return openai.ErrorWrapper(err, "do_request_failed", http.StatusInternalServerError)
 	}
 
-	err = req.Body.Close()
-	if err != nil {
-		return openai.ErrorWrapper(err, "close_request_body_failed", http.StatusInternalServerError)
-	}
-	err = c.Request.Body.Close()
-	if err != nil {
-		return openai.ErrorWrapper(err, "close_request_body_failed", http.StatusInternalServerError)
-	}
-	var imageResponse openai.ImageResponse
-
 	defer func(ctx context.Context) {
 		if resp.StatusCode != http.StatusOK {
 			return
@@ -139,34 +129,12 @@ func RelayImageHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		}
 	}(c.Request.Context())
 
-	responseBody, err := io.ReadAll(resp.Body)
-
-	if err != nil {
-		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError)
-	}
-	err = resp.Body.Close()
-	if err != nil {
-		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError)
-	}
-	err = json.Unmarshal(responseBody, &imageResponse)
-	if err != nil {
-		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError)
+	// do response
+	_, respErr := adaptor.DoResponse(c, resp, meta)
+	if respErr != nil {
+		logger.Errorf(ctx, "respErr is not nil: %+v", respErr)
+		return respErr
 	}
 
-	resp.Body = io.NopCloser(bytes.NewBuffer(responseBody))
-
-	for k, v := range resp.Header {
-		c.Writer.Header().Set(k, v[0])
-	}
-	c.Writer.WriteHeader(resp.StatusCode)
-
-	_, err = io.Copy(c.Writer, resp.Body)
-	if err != nil {
-		return openai.ErrorWrapper(err, "copy_response_body_failed", http.StatusInternalServerError)
-	}
-	err = resp.Body.Close()
-	if err != nil {
-		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError)
-	}
 	return nil
 }

relay/model/general.go

@@ -5,25 +5,29 @@ type ResponseFormat struct {
 }
 
 type GeneralOpenAIRequest struct {
-	Model            string          `json:"model,omitempty"`
 	Messages         []Message       `json:"messages,omitempty"`
-	Prompt           any             `json:"prompt,omitempty"`
-	Stream           bool            `json:"stream,omitempty"`
-	MaxTokens        int             `json:"max_tokens,omitempty"`
-	Temperature      float64         `json:"temperature,omitempty"`
-	TopP             float64         `json:"top_p,omitempty"`
-	N                int             `json:"n,omitempty"`
-	Input            any             `json:"input,omitempty"`
-	Instruction      string          `json:"instruction,omitempty"`
-	Size             string          `json:"size,omitempty"`
-	Functions        any             `json:"functions,omitempty"`
+	Model            string          `json:"model,omitempty"`
 	FrequencyPenalty float64         `json:"frequency_penalty,omitempty"`
+	MaxTokens        int             `json:"max_tokens,omitempty"`
+	N                int             `json:"n,omitempty"`
 	PresencePenalty  float64         `json:"presence_penalty,omitempty"`
 	ResponseFormat   *ResponseFormat `json:"response_format,omitempty"`
 	Seed             float64         `json:"seed,omitempty"`
-	Tools            any             `json:"tools,omitempty"`
+	Stream           bool            `json:"stream,omitempty"`
+	Temperature      float64         `json:"temperature,omitempty"`
+	TopP             float64         `json:"top_p,omitempty"`
+	TopK             int             `json:"top_k,omitempty"`
+	Tools            []Tool          `json:"tools,omitempty"`
 	ToolChoice       any             `json:"tool_choice,omitempty"`
+	FunctionCall     any             `json:"function_call,omitempty"`
+	Functions        any             `json:"functions,omitempty"`
 	User             string          `json:"user,omitempty"`
+	Prompt           any             `json:"prompt,omitempty"`
+	Input            any             `json:"input,omitempty"`
+	EncodingFormat   string          `json:"encoding_format,omitempty"`
+	Dimensions       int             `json:"dimensions,omitempty"`
+	Instruction      string          `json:"instruction,omitempty"`
+	Size             string          `json:"size,omitempty"`
 }
 
 func (r GeneralOpenAIRequest) ParseInput() []string {

relay/model/image.go

@@ -0,0 +1,12 @@
+package model
+
+type ImageRequest struct {
+	Model          string `json:"model"`
+	Prompt         string `json:"prompt" binding:"required"`
+	N              int    `json:"n,omitempty"`
+	Size           string `json:"size,omitempty"`
+	Quality        string `json:"quality,omitempty"`
+	ResponseFormat string `json:"response_format,omitempty"`
+	Style          string `json:"style,omitempty"`
+	User           string `json:"user,omitempty"`
+}

relay/model/message.go

@@ -1,9 +1,10 @@
 package model
 
 type Message struct {
-	Role    string  `json:"role"`
-	Content any     `json:"content"`
-	Name    *string `json:"name,omitempty"`
+	Role      string  `json:"role,omitempty"`
+	Content   any     `json:"content,omitempty"`
+	Name      *string `json:"name,omitempty"`
+	ToolCalls []Tool  `json:"tool_calls,omitempty"`
 }
 
 func (m Message) IsStringContent() bool {

relay/model/tool.go

@@ -0,0 +1,14 @@
+package model
+
+type Tool struct {
+	Id       string   `json:"id,omitempty"`
+	Type     string   `json:"type"`
+	Function Function `json:"function"`
+}
+
+type Function struct {
+	Description string `json:"description,omitempty"`
+	Name        string `json:"name"`
+	Parameters  any    `json:"parameters,omitempty"` // request
+	Arguments   any    `json:"arguments,omitempty"`  // response
+}

relay/util/common.go

@@ -46,6 +46,15 @@ func ShouldDisableChannel(err *relaymodel.Error, statusCode int) bool {
 	} else if strings.HasPrefix(err.Message, "This organization has been disabled.") {
 		return true
 	}
+	//if strings.Contains(err.Message, "quota") {
+	//	return true
+	//}
+	if strings.Contains(err.Message, "credit") {
+		return true
+	}
+	if strings.Contains(err.Message, "balance") {
+		return true
+	}
 	return false
 }
 

router/api-router.go

@@ -2,6 +2,7 @@ package router
 
 import (
 	"github.com/songquanpeng/one-api/controller"
+	"github.com/songquanpeng/one-api/controller/auth"
 	"github.com/songquanpeng/one-api/middleware"
 
 	"github.com/gin-contrib/gzip"
@@ -21,11 +22,13 @@ func SetApiRouter(router *gin.Engine) {
 		apiRouter.GET("/verification", middleware.CriticalRateLimit(), middleware.TurnstileCheck(), controller.SendEmailVerification)
 		apiRouter.GET("/reset_password", middleware.CriticalRateLimit(), middleware.TurnstileCheck(), controller.SendPasswordResetEmail)
 		apiRouter.POST("/user/reset", middleware.CriticalRateLimit(), controller.ResetPassword)
-		apiRouter.GET("/oauth/github", middleware.CriticalRateLimit(), controller.GitHubOAuth)
-		apiRouter.GET("/oauth/state", middleware.CriticalRateLimit(), controller.GenerateOAuthCode)
-		apiRouter.GET("/oauth/wechat", middleware.CriticalRateLimit(), controller.WeChatAuth)
-		apiRouter.GET("/oauth/wechat/bind", middleware.CriticalRateLimit(), middleware.UserAuth(), controller.WeChatBind)
+		apiRouter.GET("/oauth/github", middleware.CriticalRateLimit(), auth.GitHubOAuth)
+		apiRouter.GET("/oauth/lark", middleware.CriticalRateLimit(), auth.LarkOAuth)
+		apiRouter.GET("/oauth/state", middleware.CriticalRateLimit(), auth.GenerateOAuthCode)
+		apiRouter.GET("/oauth/wechat", middleware.CriticalRateLimit(), auth.WeChatAuth)
+		apiRouter.GET("/oauth/wechat/bind", middleware.CriticalRateLimit(), middleware.UserAuth(), auth.WeChatBind)
 		apiRouter.GET("/oauth/email/bind", middleware.CriticalRateLimit(), middleware.UserAuth(), controller.EmailBind)
+		apiRouter.POST("/topup", middleware.AdminAuth(), controller.AdminTopUp)
 
 		userRoute := apiRouter.Group("/user")
 		{
@@ -43,6 +46,7 @@ func SetApiRouter(router *gin.Engine) {
 				selfRoute.GET("/token", controller.GenerateAccessToken)
 				selfRoute.GET("/aff", controller.GetAffCode)
 				selfRoute.POST("/topup", controller.TopUp)
+				selfRoute.GET("/available_models", controller.GetUserAvailableModels)
 			}
 
 			adminRoute := userRoute.Group("/")
@@ -68,7 +72,7 @@ func SetApiRouter(router *gin.Engine) {
 		{
 			channelRoute.GET("/", controller.GetAllChannels)
 			channelRoute.GET("/search", controller.SearchChannels)
-			channelRoute.GET("/models", controller.ListModels)
+			channelRoute.GET("/models", controller.ListAllModels)
 			channelRoute.GET("/:id", controller.GetChannel)
 			channelRoute.GET("/test", controller.TestChannels)
 			channelRoute.GET("/test/:id", controller.TestChannel)

web/README.md

@@ -2,6 +2,9 @@
 
 > 每个文件夹代表一个主题，欢迎提交你的主题
 
+> [!WARNING]
+> 不是每一个主题都及时同步了所有功能，由于精力有限，优先更新默认主题，其他主题欢迎 & 期待 PR
+
 ## 提交新的主题
 
 > 欢迎在页面底部保留你和 One API 的版权信息以及指向链接
@@ -9,7 +12,7 @@
 1. 在 `web` 文件夹下新建一个文件夹，文件夹名为主题名。
 2. 把你的主题文件放到这个文件夹下。
 3. 修改你的 `package.json` 文件，把 `build` 命令改为：`"build": "react-scripts build && mv -f build ../build/default"`，其中 `default` 为你的主题名。
-4. 修改 `common/constants.go` 中的 `ValidThemes`，把你的主题名称注册进去。
+4. 修改 `common/config/config.go` 中的 `ValidThemes`，把你的主题名称注册进去。
 5. 修改 `web/THEMES` 文件，这里也需要同步修改。
 
 ## 主题列表
@@ -33,6 +36,12 @@
 |![image](https://github.com/songquanpeng/one-api/assets/42402987/fb2b1c64-ef24-4027-9b80-0cd9d945a47f)|![image](https://github.com/songquanpeng/one-api/assets/42402987/b6b649ec-2888-4324-8b2d-d5e11554eed6)|
 |![image](https://github.com/songquanpeng/one-api/assets/42402987/6d3b22e0-436b-4e26-8911-bcc993c6a2bd)|![image](https://github.com/songquanpeng/one-api/assets/42402987/eef1e224-7245-44d7-804e-9d1c8fa3f29c)|
 
+### 主题：air
+由 [Calon](https://github.com/Calcium-Ion) 开发。
+|![image](https://github.com/songquanpeng/songquanpeng.github.io/assets/39998050/1ddb274b-a715-4e81-858b-857d520b6ff4)|![image](https://github.com/songquanpeng/songquanpeng.github.io/assets/39998050/163b0b8e-1f73-49cb-b632-3dcb986b56d5)|
+|:---:|:---:|
+
+
 #### 开发说明
 
 请查看 [web/berry/README.md](https://github.com/songquanpeng/one-api/tree/main/web/berry/README.md)

web/THEMES

@@ -1,2 +1,3 @@
 default
 berry
+air

web/air/.gitignore

@@ -0,0 +1,26 @@
+# See https://help.github.com/articles/ignoring-files/ for more about ignoring files.
+
+# dependencies
+/node_modules
+/.pnp
+.pnp.js
+
+# testing
+/coverage
+
+# production
+/build
+
+# misc
+.DS_Store
+.env.local
+.env.development.local
+.env.test.local
+.env.production.local
+
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+.idea
+package-lock.json
+yarn.lock

web/air/README.md

@@ -0,0 +1,21 @@
+# React Template
+
+## Basic Usages
+
+```shell
+# Runs the app in the development mode
+npm start
+
+# Builds the app for production to the `build` folder
+npm run build
+```
+
+If you want to change the default server, please set `REACT_APP_SERVER` environment variables before build,
+for example: `REACT_APP_SERVER=http://your.domain.com`.
+
+Before you start editing, make sure your `Actions on Save` options have `Optimize imports` & `Run Prettier` enabled.
+
+## Reference
+
+1. https://github.com/OIerDb-ng/OIerDb
+2. https://github.com/cornflourblue/react-hooks-redux-registration-login-example

web/air/package.json

@@ -0,0 +1,60 @@
+{
+  "name": "react-template",
+  "version": "0.1.0",
+  "private": true,
+  "dependencies": {
+    "@douyinfe/semi-icons": "^2.46.1",
+    "@douyinfe/semi-ui": "^2.46.1",
+    "@visactor/react-vchart": "~1.8.8",
+    "@visactor/vchart": "~1.8.8",
+    "@visactor/vchart-semi-theme": "~1.8.8",
+    "axios": "^0.27.2",
+    "history": "^5.3.0",
+    "marked": "^4.1.1",
+    "react": "^18.2.0",
+    "react-dom": "^18.2.0",
+    "react-dropzone": "^14.2.3",
+    "react-fireworks": "^1.0.4",
+    "react-router-dom": "^6.3.0",
+    "react-scripts": "5.0.1",
+    "react-telegram-login": "^1.1.2",
+    "react-toastify": "^9.0.8",
+    "react-turnstile": "^1.0.5",
+    "semantic-ui-css": "^2.5.0",
+    "semantic-ui-react": "^2.1.3",
+    "usehooks-ts": "^2.9.1"
+  },
+  "scripts": {
+    "start": "react-scripts start",
+    "build": "react-scripts build && mv -f build ../build/air",
+    "test": "react-scripts test",
+    "eject": "react-scripts eject"
+  },
+  "eslintConfig": {
+    "extends": [
+      "react-app",
+      "react-app/jest"
+    ]
+  },
+  "browserslist": {
+    "production": [
+      ">0.2%",
+      "not dead",
+      "not op_mini all"
+    ],
+    "development": [
+      "last 1 chrome version",
+      "last 1 firefox version",
+      "last 1 safari version"
+    ]
+  },
+  "devDependencies": {
+    "prettier": "2.8.8",
+    "typescript": "4.4.2"
+  },
+  "prettier": {
+    "singleQuote": true,
+    "jsxSingleQuote": true
+  },
+  "proxy": "http://localhost:3000"
+}

web/air/public/index.html

@@ -0,0 +1,18 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="utf-8" />
+  <link rel="icon" href="logo.png" />
+  <meta name="viewport" content="width=device-width, initial-scale=1" />
+  <meta name="theme-color" content="#ffffff" />
+  <meta
+          name="description"
+          content="OpenAI 接口聚合管理，支持多种渠道包括 Azure，可用于二次分发管理 key，仅单可执行文件，已打包好 Docker 镜像，一键部署，开箱即用"
+  />
+  <title>One API</title>
+</head>
+<body>
+<noscript>You need to enable JavaScript to run this app.</noscript>
+<div id="root"></div>
+</body>
+</html>

web/air/public/robots.txt

@@ -0,0 +1,3 @@
+# https://www.robotstxt.org/robotstxt.html
+User-agent: *
+Disallow:

web/air/src/App.js

@@ -0,0 +1,242 @@
+import React, { lazy, Suspense, useContext, useEffect } from 'react';
+import { Route, Routes } from 'react-router-dom';
+import Loading from './components/Loading';
+import User from './pages/User';
+import { PrivateRoute } from './components/PrivateRoute';
+import RegisterForm from './components/RegisterForm';
+import LoginForm from './components/LoginForm';
+import NotFound from './pages/NotFound';
+import Setting from './pages/Setting';
+import EditUser from './pages/User/EditUser';
+import { getLogo, getSystemName } from './helpers';
+import PasswordResetForm from './components/PasswordResetForm';
+import GitHubOAuth from './components/GitHubOAuth';
+import PasswordResetConfirm from './components/PasswordResetConfirm';
+import { UserContext } from './context/User';
+import Channel from './pages/Channel';
+import Token from './pages/Token';
+import EditChannel from './pages/Channel/EditChannel';
+import Redemption from './pages/Redemption';
+import TopUp from './pages/TopUp';
+import Log from './pages/Log';
+import Chat from './pages/Chat';
+import { Layout } from '@douyinfe/semi-ui';
+import Midjourney from './pages/Midjourney';
+import Detail from './pages/Detail';
+
+const Home = lazy(() => import('./pages/Home'));
+const About = lazy(() => import('./pages/About'));
+
+function App() {
+  const [userState, userDispatch] = useContext(UserContext);
+  // const [statusState, statusDispatch] = useContext(StatusContext);
+
+  const loadUser = () => {
+    let user = localStorage.getItem('user');
+    if (user) {
+      let data = JSON.parse(user);
+      userDispatch({ type: 'login', payload: data });
+    }
+  };
+
+  useEffect(() => {
+    loadUser();
+    let systemName = getSystemName();
+    if (systemName) {
+      document.title = systemName;
+    }
+    let logo = getLogo();
+    if (logo) {
+      let linkElement = document.querySelector('link[rel~=\'icon\']');
+      if (linkElement) {
+        linkElement.href = logo;
+      }
+    }
+  }, []);
+
+  return (
+    <Layout>
+      <Layout.Content>
+        <Routes>
+          <Route
+            path="/"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <Home />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/channel"
+            element={
+              <PrivateRoute>
+                <Channel />
+              </PrivateRoute>
+            }
+          />
+          <Route
+            path="/channel/edit/:id"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <EditChannel />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/channel/add"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <EditChannel />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/token"
+            element={
+              <PrivateRoute>
+                <Token />
+              </PrivateRoute>
+            }
+          />
+          <Route
+            path="/redemption"
+            element={
+              <PrivateRoute>
+                <Redemption />
+              </PrivateRoute>
+            }
+          />
+          <Route
+            path="/user"
+            element={
+              <PrivateRoute>
+                <User />
+              </PrivateRoute>
+            }
+          />
+          <Route
+            path="/user/edit/:id"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <EditUser />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/user/edit"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <EditUser />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/user/reset"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <PasswordResetConfirm />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/login"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <LoginForm />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/register"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <RegisterForm />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/reset"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <PasswordResetForm />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/oauth/github"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <GitHubOAuth />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/setting"
+            element={
+              <PrivateRoute>
+                <Suspense fallback={<Loading></Loading>}>
+                  <Setting />
+                </Suspense>
+              </PrivateRoute>
+            }
+          />
+          <Route
+            path="/topup"
+            element={
+              <PrivateRoute>
+                <Suspense fallback={<Loading></Loading>}>
+                  <TopUp />
+                </Suspense>
+              </PrivateRoute>
+            }
+          />
+          <Route
+            path="/log"
+            element={
+              <PrivateRoute>
+                <Log />
+              </PrivateRoute>
+            }
+          />
+          <Route
+            path="/detail"
+            element={
+              <PrivateRoute>
+                <Detail />
+              </PrivateRoute>
+            }
+          />
+          <Route
+            path="/midjourney"
+            element={
+              <PrivateRoute>
+                <Midjourney />
+              </PrivateRoute>
+            }
+          />
+          <Route
+            path="/about"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <About />
+              </Suspense>
+            }
+          />
+          <Route
+            path="/chat"
+            element={
+              <Suspense fallback={<Loading></Loading>}>
+                <Chat />
+              </Suspense>
+            }
+          />
+          <Route path="*" element={
+            <NotFound />
+          } />
+        </Routes>
+      </Layout.Content>
+    </Layout>
+  );
+}
+
+export default App;

web/air/src/components/ChannelsTable.js

@@ -0,0 +1,738 @@
+import React, { useEffect, useState } from 'react';
+import { API, isMobile, shouldShowPrompt, showError, showInfo, showSuccess, timestamp2string } from '../helpers';
+
+import { CHANNEL_OPTIONS, ITEMS_PER_PAGE } from '../constants';
+import { renderGroup, renderNumberWithPoint, renderQuota } from '../helpers/render';
+import {
+  Button,
+  Dropdown,
+  Form,
+  InputNumber,
+  Popconfirm,
+  Space,
+  SplitButtonGroup,
+  Switch,
+  Table,
+  Tag,
+  Tooltip,
+  Typography
+} from '@douyinfe/semi-ui';
+import EditChannel from '../pages/Channel/EditChannel';
+import { IconTreeTriangleDown } from '@douyinfe/semi-icons';
+
+function renderTimestamp(timestamp) {
+  return (
+    <>
+      {timestamp2string(timestamp)}
+    </>
+  );
+}
+
+let type2label = undefined;
+
+function renderType(type) {
+  if (!type2label) {
+    type2label = new Map();
+    for (let i = 0; i < CHANNEL_OPTIONS.length; i++) {
+      type2label[CHANNEL_OPTIONS[i].value] = CHANNEL_OPTIONS[i];
+    }
+    type2label[0] = { value: 0, text: '未知类型', color: 'grey' };
+  }
+  return <Tag size="large" color={type2label[type]?.color}>{type2label[type]?.text}</Tag>;
+}
+
+const ChannelsTable = () => {
+  const columns = [
+    // {
+    //     title: '',
+    //     dataIndex: 'checkbox',
+    //     className: 'checkbox',
+    // },
+    {
+      title: 'ID',
+      dataIndex: 'id'
+    },
+    {
+      title: '名称',
+      dataIndex: 'name'
+    },
+    // {
+    //   title: '分组',
+    //   dataIndex: 'group',
+    //   render: (text, record, index) => {
+    //     return (
+    //       <div>
+    //         <Space spacing={2}>
+    //           {
+    //             text.split(',').map((item, index) => {
+    //               return (renderGroup(item));
+    //             })
+    //           }
+    //         </Space>
+    //       </div>
+    //     );
+    //   }
+    // },
+    {
+      title: '类型',
+      dataIndex: 'type',
+      render: (text, record, index) => {
+        return (
+          <div>
+            {renderType(text)}
+          </div>
+        );
+      }
+    },
+    {
+      title: '状态',
+      dataIndex: 'status',
+      render: (text, record, index) => {
+        return (
+          <div>
+            {renderStatus(text)}
+          </div>
+        );
+      }
+    },
+    {
+      title: '响应时间',
+      dataIndex: 'response_time',
+      render: (text, record, index) => {
+        return (
+          <div>
+            {renderResponseTime(text)}
+          </div>
+        );
+      }
+    },
+    {
+      title: '已用/剩余',
+      dataIndex: 'expired_time',
+      render: (text, record, index) => {
+        return (
+          <div>
+            <Space spacing={1}>
+              <Tooltip content={'已用额度'}>
+                <Tag color="white" type="ghost" size="large">{renderQuota(record.used_quota)}</Tag>
+              </Tooltip>
+              <Tooltip content={'剩余额度' + record.balance + '，点击更新'}>
+                <Tag color="white" type="ghost" size="large" onClick={() => {
+                  updateChannelBalance(record);
+                }}>${renderNumberWithPoint(record.balance)}</Tag>
+              </Tooltip>
+            </Space>
+          </div>
+        );
+      }
+    },
+    {
+      title: '优先级',
+      dataIndex: 'priority',
+      render: (text, record, index) => {
+        return (
+          <div>
+            <InputNumber
+              style={{ width: 70 }}
+              name="priority"
+              onBlur={e => {
+                manageChannel(record.id, 'priority', record, e.target.value);
+              }}
+              keepFocus={true}
+              innerButtons
+              defaultValue={record.priority}
+              min={-999}
+            />
+          </div>
+        );
+      }
+    },
+    // {
+    //   title: '权重',
+    //   dataIndex: 'weight',
+    //   render: (text, record, index) => {
+    //     return (
+    //       <div>
+    //         <InputNumber
+    //           style={{ width: 70 }}
+    //           name="weight"
+    //           onBlur={e => {
+    //             manageChannel(record.id, 'weight', record, e.target.value);
+    //           }}
+    //           keepFocus={true}
+    //           innerButtons
+    //           defaultValue={record.weight}
+    //           min={0}
+    //         />
+    //       </div>
+    //     );
+    //   }
+    // },
+    {
+      title: '',
+      dataIndex: 'operate',
+      render: (text, record, index) => (
+        <div>
+          {/* <SplitButtonGroup style={{ marginRight: 1 }} aria-label="测试操作项目组">
+            <Button theme="light" onClick={() => {
+              testChannel(record, '');
+            }}>测试</Button>
+            <Dropdown trigger="click" position="bottomRight" menu={record.test_models}
+            >
+              <Button style={{ padding: '8px 4px' }} type="primary" icon={<IconTreeTriangleDown />}></Button>
+            </Dropdown>
+          </SplitButtonGroup> */}
+          <Button theme='light' type='primary' style={{ marginRight: 1 }} onClick={() => testChannel(record)}>测试</Button>
+          <Popconfirm
+            title="确定是否要删除此渠道？"
+            content="此修改将不可逆"
+            okType={'danger'}
+            position={'left'}
+            onConfirm={() => {
+              manageChannel(record.id, 'delete', record).then(
+                () => {
+                  removeRecord(record.id);
+                }
+              );
+            }}
+          >
+            <Button theme="light" type="danger" style={{ marginRight: 1 }}>删除</Button>
+          </Popconfirm>
+          {
+            record.status === 1 ?
+              <Button theme="light" type="warning" style={{ marginRight: 1 }} onClick={
+                async () => {
+                  manageChannel(
+                    record.id,
+                    'disable',
+                    record
+                  );
+                }
+              }>禁用</Button> :
+              <Button theme="light" type="secondary" style={{ marginRight: 1 }} onClick={
+                async () => {
+                  manageChannel(
+                    record.id,
+                    'enable',
+                    record
+                  );
+                }
+              }>启用</Button>
+          }
+          <Button theme="light" type="tertiary" style={{ marginRight: 1 }} onClick={
+            () => {
+              setEditingChannel(record);
+              setShowEdit(true);
+            }
+          }>编辑</Button>
+        </div>
+      )
+    }
+  ];
+
+  const [channels, setChannels] = useState([]);
+  const [loading, setLoading] = useState(true);
+  const [activePage, setActivePage] = useState(1);
+  const [idSort, setIdSort] = useState(false);
+  const [searchKeyword, setSearchKeyword] = useState('');
+  const [searchGroup, setSearchGroup] = useState('');
+  const [searchModel, setSearchModel] = useState('');
+  const [searching, setSearching] = useState(false);
+  const [updatingBalance, setUpdatingBalance] = useState(false);
+  const [pageSize, setPageSize] = useState(ITEMS_PER_PAGE);
+  const [showPrompt, setShowPrompt] = useState(shouldShowPrompt('channel-test'));
+  const [channelCount, setChannelCount] = useState(pageSize);
+  const [groupOptions, setGroupOptions] = useState([]);
+  const [showEdit, setShowEdit] = useState(false);
+  const [enableBatchDelete, setEnableBatchDelete] = useState(false);
+  const [editingChannel, setEditingChannel] = useState({
+    id: undefined
+  });
+  const [selectedChannels, setSelectedChannels] = useState([]);
+
+  const removeRecord = id => {
+    let newDataSource = [...channels];
+    if (id != null) {
+      let idx = newDataSource.findIndex(data => data.id === id);
+
+      if (idx > -1) {
+        newDataSource.splice(idx, 1);
+        setChannels(newDataSource);
+      }
+    }
+  };
+
+  const setChannelFormat = (channels) => {
+    for (let i = 0; i < channels.length; i++) {
+      channels[i].key = '' + channels[i].id;
+      let test_models = [];
+      channels[i].models.split(',').forEach((item, index) => {
+        test_models.push({
+          node: 'item',
+          name: item,
+          onClick: () => {
+            testChannel(channels[i], item);
+          }
+        });
+      });
+      channels[i].test_models = test_models;
+    }
+    // data.key = '' + data.id
+    setChannels(channels);
+    if (channels.length >= pageSize) {
+      setChannelCount(channels.length + pageSize);
+    } else {
+      setChannelCount(channels.length);
+    }
+  };
+
+  const loadChannels = async (startIdx, pageSize, idSort) => {
+    setLoading(true);
+    const res = await API.get(`/api/channel/?p=${startIdx}&page_size=${pageSize}&id_sort=${idSort}`);
+    const { success, message, data } = res.data;
+    if (success) {
+      if (startIdx === 0) {
+        setChannelFormat(data);
+      } else {
+        let newChannels = [...channels];
+        newChannels.splice(startIdx * pageSize, data.length, ...data);
+        setChannelFormat(newChannels);
+      }
+    } else {
+      showError(message);
+    }
+    setLoading(false);
+  };
+
+  const refresh = async () => {
+    await loadChannels(activePage - 1, pageSize, idSort);
+  };
+
+  useEffect(() => {
+    // console.log('default effect')
+    const localIdSort = localStorage.getItem('id-sort') === 'true';
+    const localPageSize = parseInt(localStorage.getItem('page-size')) || ITEMS_PER_PAGE;
+    setIdSort(localIdSort);
+    setPageSize(localPageSize);
+    loadChannels(0, localPageSize, localIdSort)
+      .then()
+      .catch((reason) => {
+        showError(reason);
+      });
+    fetchGroups().then();
+  }, []);
+
+  const manageChannel = async (id, action, record, value) => {
+    let data = { id };
+    let res;
+    switch (action) {
+      case 'delete':
+        res = await API.delete(`/api/channel/${id}/`);
+        break;
+      case 'enable':
+        data.status = 1;
+        res = await API.put('/api/channel/', data);
+        break;
+      case 'disable':
+        data.status = 2;
+        res = await API.put('/api/channel/', data);
+        break;
+      case 'priority':
+        if (value === '') {
+          return;
+        }
+        data.priority = parseInt(value);
+        res = await API.put('/api/channel/', data);
+        break;
+      case 'weight':
+        if (value === '') {
+          return;
+        }
+        data.weight = parseInt(value);
+        if (data.weight < 0) {
+          data.weight = 0;
+        }
+        res = await API.put('/api/channel/', data);
+        break;
+    }
+    const { success, message } = res.data;
+    if (success) {
+      showSuccess('操作成功完成！');
+      let channel = res.data.data;
+      let newChannels = [...channels];
+      if (action === 'delete') {
+
+      } else {
+        record.status = channel.status;
+      }
+      setChannels(newChannels);
+    } else {
+      showError(message);
+    }
+  };
+
+  const renderStatus = (status) => {
+    switch (status) {
+      case 1:
+        return <Tag size="large" color="green">已启用</Tag>;
+      case 2:
+        return (
+          <Tag size="large" color="yellow">
+            已禁用
+          </Tag>
+        );
+      case 3:
+        return (
+          <Tag size="large" color="yellow">
+            自动禁用
+          </Tag>
+        );
+      default:
+        return (
+          <Tag size="large" color="grey">
+            未知状态
+          </Tag>
+        );
+    }
+  };
+
+  const renderResponseTime = (responseTime) => {
+    let time = responseTime / 1000;
+    time = time.toFixed(2) + ' 秒';
+    if (responseTime === 0) {
+      return <Tag size="large" color="grey">未测试</Tag>;
+    } else if (responseTime <= 1000) {
+      return <Tag size="large" color="green">{time}</Tag>;
+    } else if (responseTime <= 3000) {
+      return <Tag size="large" color="lime">{time}</Tag>;
+    } else if (responseTime <= 5000) {
+      return <Tag size="large" color="yellow">{time}</Tag>;
+    } else {
+      return <Tag size="large" color="red">{time}</Tag>;
+    }
+  };
+
+  const searchChannels = async (searchKeyword, searchGroup, searchModel) => {
+    if (searchKeyword === '' && searchGroup === '' && searchModel === '') {
+      // if keyword is blank, load files instead.
+      await loadChannels(0, pageSize, idSort);
+      setActivePage(1);
+      return;
+    }
+    setSearching(true);
+    const res = await API.get(`/api/channel/search?keyword=${searchKeyword}&group=${searchGroup}&model=${searchModel}`);
+    const { success, message, data } = res.data;
+    if (success) {
+      setChannels(data);
+      setActivePage(1);
+    } else {
+      showError(message);
+    }
+    setSearching(false);
+  };
+
+  const testChannel = async (record, model) => {
+    const res = await API.get(`/api/channel/test/${record.id}?model=${model}`);
+    const { success, message, time } = res.data;
+    if (success) {
+      record.response_time = time * 1000;
+      record.test_time = Date.now() / 1000;
+      showInfo(`渠道 ${record.name} 测试成功，耗时 ${time.toFixed(2)} 秒。`);
+    } else {
+      showError(message);
+    }
+  };
+
+  const testChannels = async (scope) => {
+    const res = await API.get(`/api/channel/test?scope=${scope}`);
+    const { success, message } = res.data;
+    if (success) {
+      showInfo('已成功开始测试渠道，请刷新页面查看结果。');
+    } else {
+      showError(message);
+    }
+  };
+
+  const deleteAllDisabledChannels = async () => {
+    const res = await API.delete(`/api/channel/disabled`);
+    const { success, message, data } = res.data;
+    if (success) {
+      showSuccess(`已删除所有禁用渠道，共计 ${data} 个`);
+      await refresh();
+    } else {
+      showError(message);
+    }
+  };
+
+  const updateChannelBalance = async (record) => {
+    const res = await API.get(`/api/channel/update_balance/${record.id}/`);
+    const { success, message, balance } = res.data;
+    if (success) {
+      record.balance = balance;
+      record.balance_updated_time = Date.now() / 1000;
+      showInfo(`渠道 ${record.name} 余额更新成功！`);
+    } else {
+      showError(message);
+    }
+  };
+
+  const updateAllChannelsBalance = async () => {
+    setUpdatingBalance(true);
+    const res = await API.get(`/api/channel/update_balance`);
+    const { success, message } = res.data;
+    if (success) {
+      showInfo('已更新完毕所有已启用渠道余额！');
+    } else {
+      showError(message);
+    }
+    setUpdatingBalance(false);
+  };
+
+  const batchDeleteChannels = async () => {
+    if (selectedChannels.length === 0) {
+      showError('请先选择要删除的渠道！');
+      return;
+    }
+    setLoading(true);
+    let ids = [];
+    selectedChannels.forEach((channel) => {
+      ids.push(channel.id);
+    });
+    const res = await API.post(`/api/channel/batch`, { ids: ids });
+    const { success, message, data } = res.data;
+    if (success) {
+      showSuccess(`已删除 ${data} 个渠道！`);
+      await refresh();
+    } else {
+      showError(message);
+    }
+    setLoading(false);
+  };
+
+  const fixChannelsAbilities = async () => {
+    const res = await API.post(`/api/channel/fix`);
+    const { success, message, data } = res.data;
+    if (success) {
+      showSuccess(`已修复 ${data} 个渠道！`);
+      await refresh();
+    } else {
+      showError(message);
+    }
+  };
+
+  let pageData = channels.slice((activePage - 1) * pageSize, activePage * pageSize);
+
+  const handlePageChange = page => {
+    setActivePage(page);
+    if (page === Math.ceil(channels.length / pageSize) + 1) {
+      // In this case we have to load more data and then append them.
+      loadChannels(page - 1, pageSize, idSort).then(r => {
+      });
+    }
+  };
+
+  const handlePageSizeChange = async (size) => {
+    localStorage.setItem('page-size', size + '');
+    setPageSize(size);
+    setActivePage(1);
+    loadChannels(0, size, idSort)
+      .then()
+      .catch((reason) => {
+        showError(reason);
+      });
+  };
+
+  const fetchGroups = async () => {
+    try {
+      let res = await API.get(`/api/group/`);
+      // add 'all' option
+      // res.data.data.unshift('all');
+      setGroupOptions(res.data.data.map((group) => ({
+        label: group,
+        value: group
+      })));
+    } catch (error) {
+      showError(error.message);
+    }
+  };
+
+  const closeEdit = () => {
+    setShowEdit(false);
+  };
+
+  const handleRow = (record, index) => {
+    if (record.status !== 1) {
+      return {
+        style: {
+          background: 'var(--semi-color-disabled-border)'
+        }
+      };
+    } else {
+      return {};
+    }
+  };
+
+
+  return (
+    <>
+      <EditChannel refresh={refresh} visible={showEdit} handleClose={closeEdit} editingChannel={editingChannel} />
+      <div style={{ display: "flex", placeItems: "center", justifyContent: "space-between" }}>
+        <Form onSubmit={() => {
+          searchChannels(searchKeyword, searchGroup, searchModel);
+        }} labelPosition="left">
+          <div style={{ display: 'flex' }}>
+            <Space>
+              <Form.Input
+                field="search_keyword"
+                label="搜索"
+                placeholder="ID，名称和密钥 ..."
+                value={searchKeyword}
+                loading={searching}
+                onChange={(v) => {
+                  setSearchKeyword(v.trim());
+                }}
+              />
+              {/* <Form.Input
+              field="search_model"
+              label="模型"
+              placeholder="模型关键字"
+              value={searchModel}
+              loading={searching}
+              onChange={(v) => {
+                setSearchModel(v.trim());
+              }}
+            />
+            <Form.Select field="group" label="分组" optionList={groupOptions} onChange={(v) => {
+              setSearchGroup(v);
+              searchChannels(searchKeyword, v, searchModel);
+            }} /> */}
+              <Button label="查询" type="primary" htmlType="submit" className="btn-margin-right"
+                style={{ marginRight: 8 }}>查询</Button>
+            </Space>
+          </div>
+        </Form>
+        <div style={{
+          display: isMobile() ? '' : 'flex',
+          marginTop: isMobile() ? 0 : -45,
+          zIndex: 999,
+          position: 'relative',
+          pointerEvents: 'none'
+        }}>
+          <Space style={{ pointerEvents: 'auto', marginTop: isMobile() ? 0 : 45 }}>
+            <Button theme="light" type="primary" style={{ marginRight: 8 }} onClick={
+              () => {
+                setEditingChannel({
+                  id: undefined
+                });
+                setShowEdit(true);
+              }
+            }>添加新的渠道</Button>
+            <Popconfirm
+              title="确定？"
+              okType={'warning'}
+              onConfirm={() => { testChannels("all") }}
+              position={isMobile() ? 'top' : 'left'}
+            >
+              <Button theme="light" type="warning" style={{ marginRight: 8 }}>测试所有渠道</Button>
+            </Popconfirm>
+            <Popconfirm
+              title="确定？"
+              okType={'warning'}
+              onConfirm={() => { testChannels("disabled") }}
+              position={isMobile() ? 'top' : 'left'}
+            >
+              <Button theme="light" type="warning" style={{ marginRight: 8 }}>测试禁用渠道</Button>
+            </Popconfirm>
+            {/* <Popconfirm
+            title="确定？"
+            okType={'secondary'}
+            onConfirm={updateAllChannelsBalance}
+          >
+            <Button theme="light" type="secondary" style={{ marginRight: 8 }}>更新所有已启用渠道余额</Button>
+          </Popconfirm> */}
+            <Popconfirm
+              title="确定是否要删除禁用渠道？"
+              content="此修改将不可逆"
+              okType={'danger'}
+              onConfirm={deleteAllDisabledChannels}
+              position={isMobile() ? 'top' : 'left'}
+            >
+              <Button theme="light" type="danger" style={{ marginRight: 8 }}>删除禁用渠道</Button>
+            </Popconfirm>
+
+            <Button theme="light" type="primary" style={{ marginRight: 8 }} onClick={refresh}>刷新</Button>
+          </Space>
+          {/*<div style={{width: '100%', pointerEvents: 'none', position: 'absolute'}}>*/}
+
+          {/*</div>*/}
+        </div>
+        {/* <div style={{ marginTop: 20 }}>
+          <Space>
+            <Typography.Text strong>开启批量删除</Typography.Text>
+            <Switch label="开启批量删除" uncheckedText="关" aria-label="是否开启批量删除" onChange={(v) => {
+              setEnableBatchDelete(v);
+            }}></Switch>
+            <Popconfirm
+              title="确定是否要删除所选渠道？"
+              content="此修改将不可逆"
+              okType={'danger'}
+              onConfirm={batchDeleteChannels}
+              disabled={!enableBatchDelete}
+              position={'top'}
+            >
+              <Button disabled={!enableBatchDelete} theme="light" type="danger"
+                style={{ marginRight: 8 }}>删除所选渠道</Button>
+            </Popconfirm>
+            <Popconfirm
+              title="确定是否要修复数据库一致性？"
+              content="进行该操作时，可能导致渠道访问错误，请仅在数据库出现问题时使用"
+              okType={'warning'}
+              onConfirm={fixChannelsAbilities}
+              position={'top'}
+            >
+              <Button theme="light" type="secondary" style={{ marginRight: 8 }}>修复数据库一致性</Button>
+            </Popconfirm>
+          </Space>
+        </div>
+        <div style={{ marginTop: 10, display: 'flex' }}>
+          <Space>
+            <Space>
+              <Typography.Text strong>使用ID排序</Typography.Text>
+              <Switch checked={idSort} label="使用ID排序" uncheckedText="关" aria-label="是否用ID排序" onChange={(v) => {
+                localStorage.setItem('id-sort', v + '');
+                setIdSort(v);
+                loadChannels(0, pageSize, v)
+                  .then()
+                  .catch((reason) => {
+                    showError(reason);
+                  });
+              }}></Switch>
+            </Space>
+          </Space>
+        </div> */}
+      </div>
+      <Table className={'channel-table'} style={{ marginTop: 15 }} columns={columns} dataSource={pageData} pagination={{
+        currentPage: activePage,
+        pageSize: pageSize,
+        total: channelCount,
+        pageSizeOpts: [10, 20, 50, 100],
+        showSizeChanger: true,
+        formatPageText: (page) => '',
+        onPageSizeChange: (size) => {
+          handlePageSizeChange(size).then();
+        },
+        onPageChange: handlePageChange
+      }} loading={loading} onRow={handleRow} rowSelection={
+        enableBatchDelete ?
+          {
+            onChange: (selectedRowKeys, selectedRows) => {
+              // console.log(`selectedRowKeys: ${selectedRowKeys}`, 'selectedRows: ', selectedRows);
+              setSelectedChannels(selectedRows);
+            }
+          } : null
+      } />
+    </>
+  );
+};
+
+export default ChannelsTable;

web/air/src/components/Footer.js

@@ -0,0 +1,64 @@
+import React, { useEffect, useState } from 'react';
+
+import { Container, Segment } from 'semantic-ui-react';
+import { getFooterHTML, getSystemName } from '../helpers';
+
+const Footer = () => {
+  const systemName = getSystemName();
+  const [footer, setFooter] = useState(getFooterHTML());
+  let remainCheckTimes = 5;
+
+  const loadFooter = () => {
+    let footer_html = localStorage.getItem('footer_html');
+    if (footer_html) {
+      setFooter(footer_html);
+    }
+  };
+
+  useEffect(() => {
+    const timer = setInterval(() => {
+      if (remainCheckTimes <= 0) {
+        clearInterval(timer);
+        return;
+      }
+      remainCheckTimes--;
+      loadFooter();
+    }, 200);
+    return () => clearTimeout(timer);
+  }, []);
+
+  return (
+    <Segment vertical>
+      <Container textAlign='center'>
+        {footer ? (
+          <div
+            className='custom-footer'
+            dangerouslySetInnerHTML={{ __html: footer }}
+          ></div>
+        ) : (
+          <div className='custom-footer'>
+            <a
+              href='https://github.com/songquanpeng/one-api'
+              target='_blank'
+            >
+              {systemName} {process.env.REACT_APP_VERSION}{' '}
+            </a>
+            由{' '}
+            <a href='https://github.com/songquanpeng' target='_blank'>
+              JustSong
+            </a>{' '}
+            构建，主题 air 来自{' '}
+            <a href='https://github.com/Calcium-Ion' target='_blank'>
+              Calon
+            </a>{' '}，源代码遵循{' '}
+            <a href='https://opensource.org/licenses/mit-license.php'>
+              MIT 协议
+            </a>
+          </div>
+        )}
+      </Container>
+    </Segment>
+  );
+};
+
+export default Footer;

web/air/src/components/GitHubOAuth.js

@@ -0,0 +1,58 @@
+import React, { useContext, useEffect, useState } from 'react';
+import { Dimmer, Loader, Segment } from 'semantic-ui-react';
+import { useNavigate, useSearchParams } from 'react-router-dom';
+import { API, showError, showSuccess } from '../helpers';
+import { UserContext } from '../context/User';
+
+const GitHubOAuth = () => {
+  const [searchParams, setSearchParams] = useSearchParams();
+
+  const [userState, userDispatch] = useContext(UserContext);
+  const [prompt, setPrompt] = useState('处理中...');
+  const [processing, setProcessing] = useState(true);
+
+  let navigate = useNavigate();
+
+  const sendCode = async (code, state, count) => {
+    const res = await API.get(`/api/oauth/github?code=${code}&state=${state}`);
+    const { success, message, data } = res.data;
+    if (success) {
+      if (message === 'bind') {
+        showSuccess('绑定成功！');
+        navigate('/setting');
+      } else {
+        userDispatch({ type: 'login', payload: data });
+        localStorage.setItem('user', JSON.stringify(data));
+        showSuccess('登录成功！');
+        navigate('/');
+      }
+    } else {
+      showError(message);
+      if (count === 0) {
+        setPrompt(`操作失败，重定向至登录界面中...`);
+        navigate('/setting'); // in case this is failed to bind GitHub
+        return;
+      }
+      count++;
+      setPrompt(`出现错误，第 ${count} 次重试中...`);
+      await new Promise((resolve) => setTimeout(resolve, count * 2000));
+      await sendCode(code, state, count);
+    }
+  };
+
+  useEffect(() => {
+    let code = searchParams.get('code');
+    let state = searchParams.get('state');
+    sendCode(code, state, 0).then();
+  }, []);
+
+  return (
+    <Segment style={{ minHeight: '300px' }}>
+      <Dimmer active inverted>
+        <Loader size="large">{prompt}</Loader>
+      </Dimmer>
+    </Segment>
+  );
+};
+
+export default GitHubOAuth;

web/air/src/components/HeaderBar.js

@@ -0,0 +1,161 @@
+import React, { useContext, useEffect, useState } from 'react';
+import { Link, useNavigate } from 'react-router-dom';
+import { UserContext } from '../context/User';
+
+import { API, getLogo, getSystemName, showSuccess } from '../helpers';
+import '../index.css';
+
+import fireworks from 'react-fireworks';
+
+import { IconHelpCircle, IconKey, IconUser } from '@douyinfe/semi-icons';
+import { Avatar, Dropdown, Layout, Nav, Switch } from '@douyinfe/semi-ui';
+import { stringToColor } from '../helpers/render';
+
+// HeaderBar Buttons
+let headerButtons = [
+  {
+    text: '关于',
+    itemKey: 'about',
+    to: '/about',
+    icon: <IconHelpCircle />
+  }
+];
+
+if (localStorage.getItem('chat_link')) {
+  headerButtons.splice(1, 0, {
+    name: '聊天',
+    to: '/chat',
+    icon: 'comments'
+  });
+}
+
+const HeaderBar = () => {
+  const [userState, userDispatch] = useContext(UserContext);
+  let navigate = useNavigate();
+
+  const [showSidebar, setShowSidebar] = useState(false);
+  const [dark, setDark] = useState(false);
+  const systemName = getSystemName();
+  const logo = getLogo();
+  var themeMode = localStorage.getItem('theme-mode');
+  const currentDate = new Date();
+  // enable fireworks on new year(1.1 and 2.9-2.24)
+  const isNewYear = (currentDate.getMonth() === 0 && currentDate.getDate() === 1) || (currentDate.getMonth() === 1 && currentDate.getDate() >= 9 && currentDate.getDate() <= 24);
+
+  async function logout() {
+    setShowSidebar(false);
+    await API.get('/api/user/logout');
+    showSuccess('注销成功!');
+    userDispatch({ type: 'logout' });
+    localStorage.removeItem('user');
+    navigate('/login');
+  }
+
+  const handleNewYearClick = () => {
+    fireworks.init('root', {});
+    fireworks.start();
+    setTimeout(() => {
+      fireworks.stop();
+      setTimeout(() => {
+        window.location.reload();
+      }, 10000);
+    }, 3000);
+  };
+
+  useEffect(() => {
+    if (themeMode === 'dark') {
+      switchMode(true);
+    }
+    if (isNewYear) {
+      console.log('Happy New Year!');
+    }
+  }, []);
+
+  const switchMode = (model) => {
+    const body = document.body;
+    if (!model) {
+      body.removeAttribute('theme-mode');
+      localStorage.setItem('theme-mode', 'light');
+    } else {
+      body.setAttribute('theme-mode', 'dark');
+      localStorage.setItem('theme-mode', 'dark');
+    }
+    setDark(model);
+  };
+  return (
+    <>
+      <Layout>
+        <div style={{ width: '100%' }}>
+          <Nav
+            mode={'horizontal'}
+            // bodyStyle={{ height: 100 }}
+            renderWrapper={({ itemElement, isSubNav, isInSubNav, props }) => {
+              const routerMap = {
+                about: '/about',
+                login: '/login',
+                register: '/register'
+              };
+              return (
+                <Link
+                  style={{ textDecoration: 'none' }}
+                  to={routerMap[props.itemKey]}
+                >
+                  {itemElement}
+                </Link>
+              );
+            }}
+            selectedKeys={[]}
+            // items={headerButtons}
+            onSelect={key => {
+
+            }}
+            footer={
+              <>
+                {isNewYear &&
+                  // happy new year
+                  <Dropdown
+                    position="bottomRight"
+                    render={
+                      <Dropdown.Menu>
+                        <Dropdown.Item onClick={handleNewYearClick}>Happy New Year!!!</Dropdown.Item>
+                      </Dropdown.Menu>
+                    }
+                  >
+                    <Nav.Item itemKey={'new-year'} text={'🏮'} />
+                  </Dropdown>
+                }
+                <Nav.Item itemKey={'about'} icon={<IconHelpCircle />} />
+                <Switch checkedText="🌞" size={'large'} checked={dark} uncheckedText="🌙" onChange={switchMode} />
+                {userState.user ?
+                  <>
+                    <Dropdown
+                      position="bottomRight"
+                      render={
+                        <Dropdown.Menu>
+                          <Dropdown.Item onClick={logout}>退出</Dropdown.Item>
+                        </Dropdown.Menu>
+                      }
+                    >
+                      <Avatar size="small" color={stringToColor(userState.user.username)} style={{ margin: 4 }}>
+                        {userState.user.username[0]}
+                      </Avatar>
+                      <span>{userState.user.username}</span>
+                    </Dropdown>
+                  </>
+                  :
+                  <>
+                    <Nav.Item itemKey={'login'} text={'登录'} icon={<IconKey />} />
+                    <Nav.Item itemKey={'register'} text={'注册'} icon={<IconUser />} />
+                  </>
+                }
+              </>
+            }
+          >
+          </Nav>
+        </div>
+      </Layout>
+    </>
+  );
+};
+
+export default HeaderBar;

web/air/src/components/Loading.js

@@ -0,0 +1,14 @@
+import React from 'react';
+import { Dimmer, Loader, Segment } from 'semantic-ui-react';
+
+const Loading = ({ prompt: name = 'page' }) => {
+  return (
+    <Segment style={{ height: 100 }}>
+      <Dimmer active inverted>
+        <Loader indeterminate>加载{name}中...</Loader>
+      </Dimmer>
+    </Segment>
+  );
+};
+
+export default Loading;

web/air/src/components/LoginForm.js

@@ -0,0 +1,254 @@
+import React, { useContext, useEffect, useState } from 'react';
+import { Link, useNavigate, useSearchParams } from 'react-router-dom';
+import { UserContext } from '../context/User';
+import { API, getLogo, showError, showInfo, showSuccess } from '../helpers';
+import { onGitHubOAuthClicked } from './utils';
+import Turnstile from 'react-turnstile';
+import { Button, Card, Divider, Form, Icon, Layout, Modal } from '@douyinfe/semi-ui';
+import Title from '@douyinfe/semi-ui/lib/es/typography/title';
+import Text from '@douyinfe/semi-ui/lib/es/typography/text';
+import TelegramLoginButton from 'react-telegram-login';
+
+import { IconGithubLogo } from '@douyinfe/semi-icons';
+import WeChatIcon from './WeChatIcon';
+
+const LoginForm = () => {
+  const [inputs, setInputs] = useState({
+    username: '',
+    password: '',
+    wechat_verification_code: ''
+  });
+  const [searchParams, setSearchParams] = useSearchParams();
+  const [submitted, setSubmitted] = useState(false);
+  const { username, password } = inputs;
+  const [userState, userDispatch] = useContext(UserContext);
+  const [turnstileEnabled, setTurnstileEnabled] = useState(false);
+  const [turnstileSiteKey, setTurnstileSiteKey] = useState('');
+  const [turnstileToken, setTurnstileToken] = useState('');
+  let navigate = useNavigate();
+  const [status, setStatus] = useState({});
+  const logo = getLogo();
+
+  useEffect(() => {
+    if (searchParams.get('expired')) {
+      showError('未登录或登录已过期，请重新登录！');
+    }
+    let status = localStorage.getItem('status');
+    if (status) {
+      status = JSON.parse(status);
+      setStatus(status);
+      if (status.turnstile_check) {
+        setTurnstileEnabled(true);
+        setTurnstileSiteKey(status.turnstile_site_key);
+      }
+    }
+  }, []);
+
+  const [showWeChatLoginModal, setShowWeChatLoginModal] = useState(false);
+
+  const onWeChatLoginClicked = () => {
+    setShowWeChatLoginModal(true);
+  };
+
+  const onSubmitWeChatVerificationCode = async () => {
+    if (turnstileEnabled && turnstileToken === '') {
+      showInfo('请稍后几秒重试，Turnstile 正在检查用户环境！');
+      return;
+    }
+    const res = await API.get(
+      `/api/oauth/wechat?code=${inputs.wechat_verification_code}`
+    );
+    const { success, message, data } = res.data;
+    if (success) {
+      userDispatch({ type: 'login', payload: data });
+      localStorage.setItem('user', JSON.stringify(data));
+      navigate('/');
+      showSuccess('登录成功！');
+      setShowWeChatLoginModal(false);
+    } else {
+      showError(message);
+    }
+  };
+
+  function handleChange(name, value) {
+    setInputs((inputs) => ({ ...inputs, [name]: value }));
+  }
+
+  async function handleSubmit(e) {
+    if (turnstileEnabled && turnstileToken === '') {
+      showInfo('请稍后几秒重试，Turnstile 正在检查用户环境！');
+      return;
+    }
+    setSubmitted(true);
+    if (username && password) {
+      const res = await API.post(`/api/user/login?turnstile=${turnstileToken}`, {
+        username,
+        password
+      });
+      const { success, message, data } = res.data;
+      if (success) {
+        userDispatch({ type: 'login', payload: data });
+        localStorage.setItem('user', JSON.stringify(data));
+        showSuccess('登录成功！');
+        if (username === 'root' && password === '123456') {
+          Modal.error({ title: '您正在使用默认密码！', content: '请立刻修改默认密码！', centered: true });
+        }
+        navigate('/token');
+      } else {
+        showError(message);
+      }
+    } else {
+      showError('请输入用户名和密码！');
+    }
+  }
+
+  // 添加Telegram登录处理函数
+  const onTelegramLoginClicked = async (response) => {
+    const fields = ['id', 'first_name', 'last_name', 'username', 'photo_url', 'auth_date', 'hash', 'lang'];
+    const params = {};
+    fields.forEach((field) => {
+      if (response[field]) {
+        params[field] = response[field];
+      }
+    });
+    const res = await API.get(`/api/oauth/telegram/login`, { params });
+    const { success, message, data } = res.data;
+    if (success) {
+      userDispatch({ type: 'login', payload: data });
+      localStorage.setItem('user', JSON.stringify(data));
+      showSuccess('登录成功！');
+      navigate('/');
+    } else {
+      showError(message);
+    }
+  };
+
+  return (
+    <div>
+      <Layout>
+        <Layout.Header>
+        </Layout.Header>
+        <Layout.Content>
+          <div style={{ justifyContent: 'center', display: 'flex', marginTop: 120 }}>
+            <div style={{ width: 500 }}>
+              <Card>
+                <Title heading={2} style={{ textAlign: 'center' }}>
+                  用户登录
+                </Title>
+                <Form>
+                  <Form.Input
+                    field={'username'}
+                    label={'用户名'}
+                    placeholder="用户名"
+                    name="username"
+                    onChange={(value) => handleChange('username', value)}
+                  />
+                  <Form.Input
+                    field={'password'}
+                    label={'密码'}
+                    placeholder="密码"
+                    name="password"
+                    type="password"
+                    onChange={(value) => handleChange('password', value)}
+                  />
+
+                  <Button theme="solid" style={{ width: '100%' }} type={'primary'} size="large"
+                          htmlType={'submit'} onClick={handleSubmit}>
+                    登录
+                  </Button>
+                </Form>
+                <div style={{ display: 'flex', justifyContent: 'space-between', marginTop: 20 }}>
+                  <Text>
+                    没有账号请先 <Link to="/register">注册账号</Link>
+                  </Text>
+                  <Text>
+                    忘记密码 <Link to="/reset">点击重置</Link>
+                  </Text>
+                </div>
+                {status.github_oauth || status.wechat_login || status.telegram_oauth ? (
+                  <>
+                    <Divider margin="12px" align="center">
+                      第三方登录
+                    </Divider>
+                    <div style={{ display: 'flex', justifyContent: 'center', marginTop: 20 }}>
+                      {status.github_oauth ? (
+                        <Button
+                          type="primary"
+                          icon={<IconGithubLogo />}
+                          onClick={() => onGitHubOAuthClicked(status.github_client_id)}
+                        />
+                      ) : (
+                        <></>
+                      )}
+                      {status.wechat_login ? (
+                        <Button
+                          type="primary"
+                          style={{ color: 'rgba(var(--semi-green-5), 1)' }}
+                          icon={<Icon svg={<WeChatIcon />} />}
+                          onClick={onWeChatLoginClicked}
+                        />
+                      ) : (
+                        <></>
+                      )}
+
+                      {status.telegram_oauth ? (
+                        <TelegramLoginButton dataOnauth={onTelegramLoginClicked} botName={status.telegram_bot_name} />
+                      ) : (
+                        <></>
+                      )}
+                    </div>
+                  </>
+                ) : (
+                  <></>
+                )}
+                <Modal
+                  title="微信扫码登录"
+                  visible={showWeChatLoginModal}
+                  maskClosable={true}
+                  onOk={onSubmitWeChatVerificationCode}
+                  onCancel={() => setShowWeChatLoginModal(false)}
+                  okText={'登录'}
+                  size={'small'}
+                  centered={true}
+                >
+                  <div style={{ display: 'flex', alignItem: 'center', flexDirection: 'column' }}>
+                    <img src={status.wechat_qrcode} />
+                  </div>
+                  <div style={{ textAlign: 'center' }}>
+                    <p>
+                      微信扫码关注公众号，输入「验证码」获取验证码（三分钟内有效）
+                    </p>
+                  </div>
+                  <Form size="large">
+                    <Form.Input
+                      field={'wechat_verification_code'}
+                      placeholder="验证码"
+                      label={'验证码'}
+                      value={inputs.wechat_verification_code}
+                      onChange={(value) => handleChange('wechat_verification_code', value)}
+                    />
+                  </Form>
+                </Modal>
+              </Card>
+              {turnstileEnabled ? (
+                <div style={{ display: 'flex', justifyContent: 'center', marginTop: 20 }}>
+                  <Turnstile
+                    sitekey={turnstileSiteKey}
+                    onVerify={(token) => {
+                      setTurnstileToken(token);
+                    }}
+                  />
+                </div>
+              ) : (
+                <></>
+              )}
+            </div>
+          </div>
+
+        </Layout.Content>
+      </Layout>
+    </div>
+  );
+};
+
+export default LoginForm;

web/air/src/components/LogsTable.js

@@ -0,0 +1,401 @@
+import React, { useEffect, useState } from 'react';
+import { API, copy, isAdmin, showError, showSuccess, timestamp2string } from '../helpers';
+
+import { Avatar, Button, Form, Layout, Modal, Select, Space, Spin, Table, Tag } from '@douyinfe/semi-ui';
+import { ITEMS_PER_PAGE } from '../constants';
+import { renderNumber, renderQuota, stringToColor } from '../helpers/render';
+import Paragraph from '@douyinfe/semi-ui/lib/es/typography/paragraph';
+
+const { Header } = Layout;
+
+function renderTimestamp(timestamp) {
+  return (<>
+    {timestamp2string(timestamp)}
+  </>);
+}
+
+const MODE_OPTIONS = [{ key: 'all', text: '全部用户', value: 'all' }, { key: 'self', text: '当前用户', value: 'self' }];
+
+const colors = ['amber', 'blue', 'cyan', 'green', 'grey', 'indigo', 'light-blue', 'lime', 'orange', 'pink', 'purple', 'red', 'teal', 'violet', 'yellow'];
+
+function renderType(type) {
+  switch (type) {
+    case 1:
+      return <Tag color="cyan" size="large"> 充值 </Tag>;
+    case 2:
+      return <Tag color="lime" size="large"> 消费 </Tag>;
+    case 3:
+      return <Tag color="orange" size="large"> 管理 </Tag>;
+    case 4:
+      return <Tag color="purple" size="large"> 系统 </Tag>;
+    default:
+      return <Tag color="black" size="large"> 未知 </Tag>;
+  }
+}
+
+function renderIsStream(bool) {
+  if (bool) {
+    return <Tag color="blue" size="large">流</Tag>;
+  } else {
+    return <Tag color="purple" size="large">非流</Tag>;
+  }
+}
+
+function renderUseTime(type) {
+  const time = parseInt(type);
+  if (time < 101) {
+    return <Tag color="green" size="large"> {time} s </Tag>;
+  } else if (time < 300) {
+    return <Tag color="orange" size="large"> {time} s </Tag>;
+  } else {
+    return <Tag color="red" size="large"> {time} s </Tag>;
+  }
+}
+
+const LogsTable = () => {
+  const columns = [{
+    title: '时间', dataIndex: 'timestamp2string'
+  }, {
+    title: '渠道',
+    dataIndex: 'channel',
+    className: isAdmin() ? 'tableShow' : 'tableHiddle',
+    render: (text, record, index) => {
+      return (isAdminUser ? record.type === 0 || record.type === 2 ? <div>
+        {<Tag color={colors[parseInt(text) % colors.length]} size="large"> {text} </Tag>}
+      </div> : <></> : <></>);
+    }
+  }, {
+    title: '用户',
+    dataIndex: 'username',
+    className: isAdmin() ? 'tableShow' : 'tableHiddle',
+    render: (text, record, index) => {
+      return (isAdminUser ? <div>
+        <Avatar size="small" color={stringToColor(text)} style={{ marginRight: 4 }}
+          onClick={() => showUserInfo(record.user_id)}>
+          {typeof text === 'string' && text.slice(0, 1)}
+        </Avatar>
+        {text}
+      </div> : <></>);
+    }
+  }, {
+    title: '令牌', dataIndex: 'token_name', render: (text, record, index) => {
+      return (record.type === 0 || record.type === 2 ? <div>
+        <Tag color="grey" size="large" onClick={() => {
+          copyText(text);
+        }}> {text} </Tag>
+      </div> : <></>);
+    }
+  }, {
+    title: '类型', dataIndex: 'type', render: (text, record, index) => {
+      return (<div>
+        {renderType(text)}
+      </div>);
+    }
+  }, {
+    title: '模型', dataIndex: 'model_name', render: (text, record, index) => {
+      return (record.type === 0 || record.type === 2 ? <div>
+        <Tag color={stringToColor(text)} size="large" onClick={() => {
+          copyText(text);
+        }}> {text} </Tag>
+      </div> : <></>);
+    }
+  },
+  // {
+  //   title: '用时', dataIndex: 'use_time', render: (text, record, index) => {
+  //     return (<div>
+  //       <Space>
+  //         {renderUseTime(text)}
+  //         {renderIsStream(record.is_stream)}
+  //       </Space>
+  //     </div>);
+  //   }
+  // },
+  {
+    title: '提示', dataIndex: 'prompt_tokens', render: (text, record, index) => {
+      return (record.type === 0 || record.type === 2 ? <div>
+        {<span> {text} </span>}
+      </div> : <></>);
+    }
+  }, {
+    title: '补全', dataIndex: 'completion_tokens', render: (text, record, index) => {
+      return (parseInt(text) > 0 && (record.type === 0 || record.type === 2) ? <div>
+        {<span> {text} </span>}
+      </div> : <></>);
+    }
+  }, {
+    title: '花费', dataIndex: 'quota', render: (text, record, index) => {
+      return (record.type === 0 || record.type === 2 ? <div>
+        {renderQuota(text, 6)}
+      </div> : <></>);
+    }
+  }, {
+    title: '详情', dataIndex: 'content', render: (text, record, index) => {
+      return <Paragraph ellipsis={{ rows: 2, showTooltip: { type: 'popover', opts: { style: { width: 240 } } } }}
+        style={{ maxWidth: 240 }}>
+        {text}
+      </Paragraph>;
+    }
+  }];
+
+  const [logs, setLogs] = useState([]);
+  const [showStat, setShowStat] = useState(false);
+  const [loading, setLoading] = useState(false);
+  const [loadingStat, setLoadingStat] = useState(false);
+  const [activePage, setActivePage] = useState(1);
+  const [logCount, setLogCount] = useState(ITEMS_PER_PAGE);
+  const [pageSize, setPageSize] = useState(ITEMS_PER_PAGE);
+  const [searchKeyword, setSearchKeyword] = useState('');
+  const [searching, setSearching] = useState(false);
+  const [logType, setLogType] = useState(0);
+  const isAdminUser = isAdmin();
+  let now = new Date();
+  // 初始化start_timestamp为前一天
+  const [inputs, setInputs] = useState({
+    username: '',
+    token_name: '',
+    model_name: '',
+    start_timestamp: timestamp2string(now.getTime() / 1000 - 86400),
+    end_timestamp: timestamp2string(now.getTime() / 1000 + 3600),
+    channel: ''
+  });
+  const { username, token_name, model_name, start_timestamp, end_timestamp, channel } = inputs;
+
+  const [stat, setStat] = useState({
+    quota: 0, token: 0
+  });
+
+  const handleInputChange = (value, name) => {
+    setInputs((inputs) => ({ ...inputs, [name]: value }));
+  };
+
+  const getLogSelfStat = async () => {
+    let localStartTimestamp = Date.parse(start_timestamp) / 1000;
+    let localEndTimestamp = Date.parse(end_timestamp) / 1000;
+    let res = await API.get(`/api/log/self/stat?type=${logType}&token_name=${token_name}&model_name=${model_name}&start_timestamp=${localStartTimestamp}&end_timestamp=${localEndTimestamp}`);
+    const { success, message, data } = res.data;
+    if (success) {
+      setStat(data);
+    } else {
+      showError(message);
+    }
+  };
+
+  const getLogStat = async () => {
+    let localStartTimestamp = Date.parse(start_timestamp) / 1000;
+    let localEndTimestamp = Date.parse(end_timestamp) / 1000;
+    let res = await API.get(`/api/log/stat?type=${logType}&username=${username}&token_name=${token_name}&model_name=${model_name}&start_timestamp=${localStartTimestamp}&end_timestamp=${localEndTimestamp}&channel=${channel}`);
+    const { success, message, data } = res.data;
+    if (success) {
+      setStat(data);
+    } else {
+      showError(message);
+    }
+  };
+
+  const handleEyeClick = async () => {
+    setLoadingStat(true);
+    if (isAdminUser) {
+      await getLogStat();
+    } else {
+      await getLogSelfStat();
+    }
+    setShowStat(true);
+    setLoadingStat(false);
+  };
+
+  const showUserInfo = async (userId) => {
+    if (!isAdminUser) {
+      return;
+    }
+    const res = await API.get(`/api/user/${userId}`);
+    const { success, message, data } = res.data;
+    if (success) {
+      Modal.info({
+        title: '用户信息', content: <div style={{ padding: 12 }}>
+          <p>用户名: {data.username}</p>
+          <p>余额: {renderQuota(data.quota)}</p>
+          <p>已用额度：{renderQuota(data.used_quota)}</p>
+          <p>请求次数：{renderNumber(data.request_count)}</p>
+        </div>, centered: true
+      });
+    } else {
+      showError(message);
+    }
+  };
+
+  const setLogsFormat = (logs) => {
+    for (let i = 0; i < logs.length; i++) {
+      logs[i].timestamp2string = timestamp2string(logs[i].created_at);
+      logs[i].key = '' + logs[i].id;
+    }
+    // data.key = '' + data.id
+    setLogs(logs);
+    setLogCount(logs.length + ITEMS_PER_PAGE);
+    // console.log(logCount);
+  };
+
+  const loadLogs = async (startIdx, pageSize, logType = 0) => {
+    setLoading(true);
+
+    let url = '';
+    let localStartTimestamp = Date.parse(start_timestamp) / 1000;
+    let localEndTimestamp = Date.parse(end_timestamp) / 1000;
+    if (isAdminUser) {
+      url = `/api/log/?p=${startIdx}&page_size=${pageSize}&type=${logType}&username=${username}&token_name=${token_name}&model_name=${model_name}&start_timestamp=${localStartTimestamp}&end_timestamp=${localEndTimestamp}&channel=${channel}`;
+    } else {
+      url = `/api/log/self/?p=${startIdx}&page_size=${pageSize}&type=${logType}&token_name=${token_name}&model_name=${model_name}&start_timestamp=${localStartTimestamp}&end_timestamp=${localEndTimestamp}`;
+    }
+    const res = await API.get(url);
+    const { success, message, data } = res.data;
+    if (success) {
+      if (startIdx === 0) {
+        setLogsFormat(data);
+      } else {
+        let newLogs = [...logs];
+        newLogs.splice(startIdx * pageSize, data.length, ...data);
+        setLogsFormat(newLogs);
+      }
+    } else {
+      showError(message);
+    }
+    setLoading(false);
+  };
+
+  const pageData = logs.slice((activePage - 1) * pageSize, activePage * pageSize);
+
+  const handlePageChange = page => {
+    setActivePage(page);
+    if (page === Math.ceil(logs.length / pageSize) + 1) {
+      // In this case we have to load more data and then append them.
+      loadLogs(page - 1, pageSize).then(r => {
+      });
+    }
+  };
+
+  const handlePageSizeChange = async (size) => {
+    localStorage.setItem('page-size', size + '');
+    setPageSize(size);
+    setActivePage(1);
+    loadLogs(0, size)
+      .then()
+      .catch((reason) => {
+        showError(reason);
+      });
+  };
+
+  const refresh = async (localLogType) => {
+    // setLoading(true);
+    setActivePage(1);
+    await loadLogs(0, pageSize, localLogType);
+  };
+
+  const copyText = async (text) => {
+    if (await copy(text)) {
+      showSuccess('已复制：' + text);
+    } else {
+      // setSearchKeyword(text);
+      Modal.error({ title: '无法复制到剪贴板，请手动复制', content: text });
+    }
+  };
+
+  useEffect(() => {
+    // console.log('default effect')
+    const localPageSize = parseInt(localStorage.getItem('page-size')) || ITEMS_PER_PAGE;
+    setPageSize(localPageSize);
+    loadLogs(0, localPageSize)
+      .then()
+      .catch((reason) => {
+        showError(reason);
+      });
+  }, []);
+
+  const searchLogs = async () => {
+    if (searchKeyword === '') {
+      // if keyword is blank, load files instead.
+      await loadLogs(0, pageSize);
+      setActivePage(1);
+      return;
+    }
+    setSearching(true);
+    const res = await API.get(`/api/log/self/search?keyword=${searchKeyword}`);
+    const { success, message, data } = res.data;
+    if (success) {
+      setLogs(data);
+      setActivePage(1);
+    } else {
+      showError(message);
+    }
+    setSearching(false);
+  };
+
+  return (<>
+    <Layout>
+      <Header>
+        <Spin spinning={loadingStat}>
+          <h3>使用明细（总消耗额度：
+            <span onClick={handleEyeClick} style={{
+              cursor: 'pointer', color: 'gray'
+            }}>{showStat ? renderQuota(stat.quota) : '点击查看'}</span>
+            ）
+          </h3>
+        </Spin>
+      </Header>
+      <Form layout="horizontal" style={{ marginTop: 10 }}>
+        <>
+          <Form.Input field="token_name" label="令牌名称" style={{ width: 176 }} value={token_name}
+            placeholder={'可选值'} name="token_name"
+            onChange={value => handleInputChange(value, 'token_name')} />
+          <Form.Input field="model_name" label="模型名称" style={{ width: 176 }} value={model_name}
+            placeholder="可选值"
+            name="model_name"
+            onChange={value => handleInputChange(value, 'model_name')} />
+          <Form.DatePicker field="start_timestamp" label="起始时间" style={{ width: 272 }}
+            initValue={start_timestamp}
+            value={start_timestamp} type="dateTime"
+            name="start_timestamp"
+            onChange={value => handleInputChange(value, 'start_timestamp')} />
+          <Form.DatePicker field="end_timestamp" fluid label="结束时间" style={{ width: 272 }}
+            initValue={end_timestamp}
+            value={end_timestamp} type="dateTime"
+            name="end_timestamp"
+            onChange={value => handleInputChange(value, 'end_timestamp')} />
+          {isAdminUser && <>
+            <Form.Input field="channel" label="渠道 ID" style={{ width: 176 }} value={channel}
+              placeholder="可选值" name="channel"
+              onChange={value => handleInputChange(value, 'channel')} />
+            <Form.Input field="username" label="用户名称" style={{ width: 176 }} value={username}
+              placeholder={'可选值'} name="username"
+              onChange={value => handleInputChange(value, 'username')} />
+          </>}
+          <Form.Section>
+            <Button label="查询" type="primary" htmlType="submit" className="btn-margin-right"
+              onClick={refresh} loading={loading}>查询</Button>
+          </Form.Section>
+        </>
+      </Form>
+      <Table style={{ marginTop: 5 }} columns={columns} dataSource={pageData} pagination={{
+        currentPage: activePage,
+        pageSize: pageSize,
+        total: logCount,
+        pageSizeOpts: [10, 20, 50, 100],
+        showSizeChanger: true,
+        onPageSizeChange: (size) => {
+          handlePageSizeChange(size).then();
+        },
+        onPageChange: handlePageChange
+      }} />
+      <Select defaultValue="0" style={{ width: 120 }} onChange={(value) => {
+        setLogType(parseInt(value));
+        refresh(parseInt(value)).then();
+      }}>
+        <Select.Option value="0">全部</Select.Option>
+        <Select.Option value="1">充值</Select.Option>
+        <Select.Option value="2">消费</Select.Option>
+        <Select.Option value="3">管理</Select.Option>
+        <Select.Option value="4">系统</Select.Option>
+      </Select>
+    </Layout>
+  </>);
+};
+
+export default LogsTable;