feat:gemini file

build:dockerfile
2025-11-20 23:16:49 +08:00 · 2025-09-15 17:16:50 +08:00 · 2025-09-15 16:18:27 +08:00 · 2025-09-15 15:58:46 +08:00 · 2025-09-15 15:33:48 +08:00 · 2025-09-15 15:30:56 +08:00
72 changed files with 1955 additions and 455 deletions
--- a/.github/workflows/docker-image.yml
+++ b/.github/workflows/docker-image.yml
@@ -32,10 +32,10 @@ jobs:
          git describe --tags > VERSION 

      - name: Set up QEMU
-        uses: docker/setup-qemu-action@v2
+        uses: docker/setup-qemu-action@v3

      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v2
+        uses: docker/setup-buildx-action@v3

      - name: Log in to Docker Hub
        uses: docker/login-action@v2
@@ -62,8 +62,7 @@ jobs:
        uses: docker/build-push-action@v3
        with:
          context: .
-          #          platforms: linux/amd64,linux/arm64
-          platforms: linux/amd64 # TODO disable arm64 for now, because it cause error
+          platforms: ${{ contains(github.ref, 'alpha') && 'linux/amd64' || 'linux/amd64' }}
          push: true
          tags: ${{ steps.meta.outputs.tags }}
          labels: ${{ steps.meta.outputs.labels }}
--- a/.github/workflows/linux-release.yml
+++ b/.github/workflows/linux-release.yml
@@ -7,6 +7,7 @@ on:
    tags:
      - 'v*.*.*'
      - '!*-alpha*'
+      - '!*-preview*'
  workflow_dispatch:
    inputs:
      name:
--- a/.github/workflows/macos-release.yml
+++ b/.github/workflows/macos-release.yml
@@ -7,6 +7,7 @@ on:
    tags:
      - 'v*.*.*'
      - '!*-alpha*'
+      - '!*-preview*'
  workflow_dispatch:
    inputs:
      name:
--- a/.github/workflows/windows-release.yml
+++ b/.github/workflows/windows-release.yml
@@ -7,6 +7,7 @@ on:
    tags:
      - 'v*.*.*'
      - '!*-alpha*'
+      - '!*-preview*'
  workflow_dispatch:
    inputs:
      name:
--- a/30
+++ b/30
@@ -9,23 +9,22 @@ RUN npm install --prefix /web/default & \
    npm install --prefix /web/air & \
    wait

-RUN DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat /web/default/VERSION) npm run build --prefix /web/default & \
-    DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat /web/berry/VERSION) npm run build --prefix /web/berry & \
-    DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat /web/air/VERSION) npm run build --prefix /web/air & \
+RUN DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat ./VERSION) npm run build --prefix /web/default & \
+    DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat ./VERSION) npm run build --prefix /web/berry & \
+    DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat ./VERSION) npm run build --prefix /web/air & \
    wait

-FROM golang AS builder2
+FROM golang:alpine AS builder2

-RUN apt-get update && apt-get install -y --no-install-recommends \
-    build-essential \
-    sqlite3 libsqlite3-dev \
-    && rm -rf /var/lib/apt/lists/*
+RUN apk add --no-cache \
+    gcc \
+    musl-dev \
+    sqlite-dev \
+    build-base

 ENV GO111MODULE=on \
    CGO_ENABLED=1 \
-    GOOS=linux \
-    CGO_CFLAGS="-I/usr/include" \
-    CGO_LDFLAGS="-L/usr/lib"
+    GOOS=linux

 WORKDIR /build

@@ -35,14 +34,11 @@ RUN go mod download
 COPY . .
 COPY --from=builder /web/build ./web/build

-RUN go build -trimpath -ldflags "-s -w -X 'github.com/songquanpeng/one-api/common.Version=$(cat VERSION)'" -o one-api
+RUN go build -trimpath -ldflags "-s -w -X 'github.com/songquanpeng/one-api/common.Version=$(cat VERSION)' -linkmode external -extldflags '-static'" -o one-api

-# Final runtime image
-FROM ubuntu:22.04
+FROM alpine:latest

-RUN apt-get update && apt-get install -y --no-install-recommends \
-    ca-certificates tzdata bash \
-    && rm -rf /var/lib/apt/lists/*
+RUN apk add --no-cache ca-certificates tzdata

 COPY --from=builder2 /build/one-api /

--- a/README.en.md
+++ b/README.en.md
@@ -315,6 +315,7 @@ If the channel ID is not provided, load balancing will be used to distribute the
 * [FastGPT](https://github.com/labring/FastGPT): Knowledge question answering system based on the LLM
 * [VChart](https://github.com/VisActor/VChart):  More than just a cross-platform charting library, but also an expressive data storyteller.
 * [VMind](https://github.com/VisActor/VMind):  Not just automatic, but also fantastic. Open-source solution for intelligent visualization.
+* * [CherryStudio](https://github.com/CherryHQ/cherry-studio):  A cross-platform AI client that integrates multiple service providers and supports local knowledge base management.

 ## Note
 This project is an open-source project. Please use it in compliance with OpenAI's [Terms of Use](https://openai.com/policies/terms-of-use) and **applicable laws and regulations**. It must not be used for illegal purposes.
--- a/README.ja.md
+++ b/README.ja.md
@@ -287,8 +287,8 @@ graph LR
    + インターフェイスアドレスと API Key が正しいか再確認してください。

 ## 関連プロジェクト
-[FastGPT](https://github.com/labring/FastGPT): LLM に基づく知識質問応答システム
-
+* [FastGPT](https://github.com/labring/FastGPT): LLM に基づく知識質問応答システム
+* [CherryStudio](https://github.com/CherryHQ/cherry-studio):  マルチプラットフォーム対応のAIクライアント。複数のサービスプロバイダーを統合管理し、ローカル知識ベースをサポートします。
 ## 注
 本プロジェクトはオープンソースプロジェクトです。OpenAI の[利用規約](https://openai.com/policies/terms-of-use)および**適用される法令**を遵守してご利用ください。違法な目的での利用はご遠慮ください。

--- a/README.md
+++ b/README.md
@@ -72,7 +72,7 @@ _✨ 通过标准的 OpenAI API 格式访问所有的大模型，开箱即用
   + [x] [Anthropic Claude 系列模型](https://anthropic.com) (支持 AWS Claude)
   + [x] [Google PaLM2/Gemini 系列模型](https://developers.generativeai.google)
   + [x] [Mistral 系列模型](https://mistral.ai/)
-   + [x] [字节跳动豆包大模型](https://console.volcengine.com/ark/region:ark+cn-beijing/model)
+   + [x] [字节跳动豆包大模型（火山引擎）](https://www.volcengine.com/experience/ark?utm_term=202502dsinvite&ac=DSASUQY5&rc=2QXCA1VI)
   + [x] [百度文心一言系列模型](https://cloud.baidu.com/doc/WENXINWORKSHOP/index.html)
   + [x] [阿里通义千问系列模型](https://help.aliyun.com/document_detail/2400395.html)
   + [x] [讯飞星火认知大模型](https://www.xfyun.cn/doc/spark/Web.html)
@@ -115,7 +115,7 @@ _✨ 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)。。
+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。
@@ -469,6 +469,7 @@ https://openai.justsong.cn
 * [ChatGPT Next Web](https://github.com/Yidadaa/ChatGPT-Next-Web):  一键拥有你自己的跨平台 ChatGPT 应用
 * [VChart](https://github.com/VisActor/VChart):  不只是开箱即用的多端图表库，更是生动灵活的数据故事讲述者。
 * [VMind](https://github.com/VisActor/VMind):  不仅自动，还很智能。开源智能可视化解决方案。
+* [CherryStudio](https://github.com/CherryHQ/cherry-studio):  全平台支持的AI客户端, 多服务商集成管理、本地知识库支持。

 ## 注意

--- a/common/config/config.go
+++ b/common/config/config.go
@@ -163,4 +163,4 @@ var UserContentRequestProxy = env.String("USER_CONTENT_REQUEST_PROXY", "")
 var UserContentRequestTimeout = env.Int("USER_CONTENT_REQUEST_TIMEOUT", 30)

 var EnforceIncludeUsage = env.Bool("ENFORCE_INCLUDE_USAGE", false)
-var TestPrompt = env.String("TEST_PROMPT", "Print your model name exactly and do not output without any other text.")
+var TestPrompt = env.String("TEST_PROMPT", "Output only your specific model name with no additional text.")
--- a/common/file/file.go
+++ b/common/file/file.go
@@ -0,0 +1,23 @@
+package file
+
+import (
+	"bytes"
+	"encoding/base64"
+	"net/http"
+)
+
+func GetFileFromUrl(url string) (mimeType string, data string, err error) {
+	resp, err := http.Get(url)
+	if err != nil {
+		return
+	}
+	defer resp.Body.Close()
+	buffer := bytes.NewBuffer(nil)
+	_, err = buffer.ReadFrom(resp.Body)
+	if err != nil {
+		return
+	}
+	mimeType = resp.Header.Get("Content-Type")
+	data = base64.StdEncoding.EncodeToString(buffer.Bytes())
+	return
+}
--- a/common/logger/logger.go
+++ b/common/logger/logger.go
@@ -93,6 +93,9 @@ func Error(ctx context.Context, msg string) {
 }

 func Debugf(ctx context.Context, format string, a ...any) {
+	if !config.DebugEnabled {
+		return
+	}
 	logHelper(ctx, loggerDEBUG, fmt.Sprintf(format, a...))
 }

--- a/common/utils/array.go
+++ b/common/utils/array.go
@@ -0,0 +1,13 @@
+package utils
+
+func DeDuplication(slice []string) []string {
+	m := make(map[string]bool)
+	for _, v := range slice {
+		m[v] = true
+	}
+	result := make([]string, 0, len(m))
+	for v := range m {
+		result = append(result, v)
+	}
+	return result
+}
--- a/controller/channel-billing.go
+++ b/controller/channel-billing.go
@@ -112,6 +112,13 @@ type DeepSeekUsageResponse struct {
 	} `json:"balance_infos"`
 }

+type OpenRouterResponse struct {
+	Data struct {
+		TotalCredits float64 `json:"total_credits"`
+		TotalUsage   float64 `json:"total_usage"`
+	} `json:"data"`
+}
+
 // GetAuthHeader get auth header
 func GetAuthHeader(token string) http.Header {
 	h := http.Header{}
@@ -285,6 +292,22 @@ func updateChannelDeepSeekBalance(channel *model.Channel) (float64, error) {
 	return balance, nil
 }

+func updateChannelOpenRouterBalance(channel *model.Channel) (float64, error) {
+	url := "https://openrouter.ai/api/v1/credits"
+	body, err := GetResponseBody("GET", url, channel, GetAuthHeader(channel.Key))
+	if err != nil {
+		return 0, err
+	}
+	response := OpenRouterResponse{}
+	err = json.Unmarshal(body, &response)
+	if err != nil {
+		return 0, err
+	}
+	balance := response.Data.TotalCredits - response.Data.TotalUsage
+	channel.UpdateBalance(balance)
+	return balance, nil
+}
+
 func updateChannelBalance(channel *model.Channel) (float64, error) {
 	baseURL := channeltype.ChannelBaseURLs[channel.Type]
 	if channel.GetBaseURL() == "" {
@@ -313,6 +336,8 @@ func updateChannelBalance(channel *model.Channel) (float64, error) {
 		return updateChannelSiliconFlowBalance(channel)
 	case channeltype.DeepSeek:
 		return updateChannelDeepSeekBalance(channel)
+	case channeltype.OpenRouter:
+		return updateChannelOpenRouterBalance(channel)
 	default:
 		return 0, errors.New("尚未实现")
 	}
--- a/controller/channel-test.go
+++ b/controller/channel-test.go
@@ -153,6 +153,7 @@ func testChannel(ctx context.Context, channel *model.Channel, request *relaymode
 	rawResponse := w.Body.String()
 	_, responseMessage, err = parseTestResponse(rawResponse)
 	if err != nil {
+		logger.SysError(fmt.Sprintf("failed to parse error: %s, \nresponse: %s", err.Error(), rawResponse))
 		return "", err, nil
 	}
 	result := w.Result()
--- a/model/ability.go
+++ b/model/ability.go
@@ -2,10 +2,13 @@ package model

 import (
 	"context"
-	"github.com/songquanpeng/one-api/common"
-	"gorm.io/gorm"
 	"sort"
 	"strings"
+
+	"gorm.io/gorm"
+
+	"github.com/songquanpeng/one-api/common"
+	"github.com/songquanpeng/one-api/common/utils"
 )

 type Ability struct {
@@ -49,6 +52,7 @@ func GetRandomSatisfiedChannel(group string, model string, ignoreFirstPriority b

 func (channel *Channel) AddAbilities() error {
 	models_ := strings.Split(channel.Models, ",")
+	models_ = utils.DeDuplication(models_)
 	groups_ := strings.Split(channel.Group, ",")
 	abilities := make([]Ability, 0, len(models_))
 	for _, model := range models_ {
--- a/model/main.go
+++ b/model/main.go
@@ -135,30 +135,32 @@ func InitDB() {
 }

 func migrateDB() error {
-	var err error
-	if err = DB.AutoMigrate(&Channel{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Token{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&User{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Option{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Redemption{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Ability{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Log{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Channel{}); err != nil {
-		return err
+	if env.Bool("StartSqlMigration", false) {
+		var err error
+		if err = DB.AutoMigrate(&Channel{}); err != nil {
+			return err
+		}
+		if err = DB.AutoMigrate(&Token{}); err != nil {
+			return err
+		}
+		if err = DB.AutoMigrate(&User{}); err != nil {
+			return err
+		}
+		if err = DB.AutoMigrate(&Option{}); err != nil {
+			return err
+		}
+		if err = DB.AutoMigrate(&Redemption{}); err != nil {
+			return err
+		}
+		if err = DB.AutoMigrate(&Ability{}); err != nil {
+			return err
+		}
+		if err = DB.AutoMigrate(&Log{}); err != nil {
+			return err
+		}
+		if err = DB.AutoMigrate(&Channel{}); err != nil {
+			return err
+		}
 	}
 	return nil
 }
--- a/relay/adaptor.go
+++ b/relay/adaptor.go
@@ -64,6 +64,9 @@ func GetAdaptor(apiType int) adaptor.Adaptor {
 		return &proxy.Adaptor{}
 	case apitype.Replicate:
 		return &replicate.Adaptor{}
+	case apitype.CozeV3:
+		return &coze.AdaptorV3{}
+
 	}
 	return nil
 }
--- a/relay/adaptor/ali/constants.go
+++ b/relay/adaptor/ali/constants.go
@@ -14,10 +14,14 @@ var ModelList = []string{
 	"qwen2-72b-instruct", "qwen2-57b-a14b-instruct", "qwen2-7b-instruct", "qwen2-1.5b-instruct", "qwen2-0.5b-instruct",
 	"qwen1.5-110b-chat", "qwen1.5-72b-chat", "qwen1.5-32b-chat", "qwen1.5-14b-chat", "qwen1.5-7b-chat", "qwen1.5-1.8b-chat", "qwen1.5-0.5b-chat",
 	"qwen-72b-chat", "qwen-14b-chat", "qwen-7b-chat", "qwen-1.8b-chat", "qwen-1.8b-longcontext-chat",
+	"qvq-72b-preview",
+	"qwen2.5-vl-72b-instruct", "qwen2.5-vl-7b-instruct", "qwen2.5-vl-2b-instruct", "qwen2.5-vl-1b-instruct", "qwen2.5-vl-0.5b-instruct",
 	"qwen2-vl-7b-instruct", "qwen2-vl-2b-instruct", "qwen-vl-v1", "qwen-vl-chat-v1",
 	"qwen2-audio-instruct", "qwen-audio-chat",
 	"qwen2.5-math-72b-instruct", "qwen2.5-math-7b-instruct", "qwen2.5-math-1.5b-instruct", "qwen2-math-72b-instruct", "qwen2-math-7b-instruct", "qwen2-math-1.5b-instruct",
 	"qwen2.5-coder-32b-instruct", "qwen2.5-coder-14b-instruct", "qwen2.5-coder-7b-instruct", "qwen2.5-coder-3b-instruct", "qwen2.5-coder-1.5b-instruct", "qwen2.5-coder-0.5b-instruct",
 	"text-embedding-v1", "text-embedding-v3", "text-embedding-v2", "text-embedding-async-v2", "text-embedding-async-v1",
 	"ali-stable-diffusion-xl", "ali-stable-diffusion-v1.5", "wanx-v1",
+	"qwen-mt-plus", "qwen-mt-turbo",
+	"deepseek-r1", "deepseek-v3", "deepseek-r1-distill-qwen-1.5b", "deepseek-r1-distill-qwen-7b", "deepseek-r1-distill-qwen-14b", "deepseek-r1-distill-qwen-32b", "deepseek-r1-distill-llama-8b", "deepseek-r1-distill-llama-70b",
 }
--- a/relay/adaptor/ali/main.go
+++ b/relay/adaptor/ali/main.go
@@ -36,6 +36,12 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
 		enableSearch = true
 		aliModel = strings.TrimSuffix(aliModel, EnableSearchModelSuffix)
 	}
+
+	enableThinking := false
+	if request.ReasoningEffort != nil {
+		enableThinking = true
+	}
+
 	request.TopP = helper.Float64PtrMax(request.TopP, 0.9999)
 	return &ChatRequest{
 		Model: aliModel,
@@ -52,6 +58,7 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
 			TopK:              request.TopK,
 			ResultFormat:      "message",
 			Tools:             request.Tools,
+			EnableThinking:    enableThinking,
 		},
 	}
 }
--- a/relay/adaptor/ali/model.go
+++ b/relay/adaptor/ali/model.go
@@ -25,6 +25,7 @@ type Parameters struct {
 	Temperature       *float64     `json:"temperature,omitempty"`
 	ResultFormat      string       `json:"result_format,omitempty"`
 	Tools             []model.Tool `json:"tools,omitempty"`
+	EnableThinking    bool         `json:"enable_thinking,omitempty"`
 }

 type ChatRequest struct {
--- a/relay/adaptor/alibailian/constants.go
+++ b/relay/adaptor/alibailian/constants.go
@@ -0,0 +1,21 @@
+package alibailian
+
+// https://help.aliyun.com/zh/model-studio/getting-started/models
+
+var ModelList = []string{
+	"qwen-turbo",
+	"qwen-plus",
+	"qwen-long",
+	"qwen-max",
+	"qwen-coder-plus",
+	"qwen-coder-plus-latest",
+	"qwen-coder-turbo",
+	"qwen-coder-turbo-latest",
+	"qwen-mt-plus",
+	"qwen-mt-turbo",
+	"qwq-32b-preview",
+
+	"deepseek-r1",
+	"deepseek-v3",
+	"deepseek-v3.1",
+}
--- a/relay/adaptor/alibailian/main.go
+++ b/relay/adaptor/alibailian/main.go
@@ -0,0 +1,19 @@
+package alibailian
+
+import (
+	"fmt"
+
+	"github.com/songquanpeng/one-api/relay/meta"
+	"github.com/songquanpeng/one-api/relay/relaymode"
+)
+
+func GetRequestURL(meta *meta.Meta) (string, error) {
+	switch meta.Mode {
+	case relaymode.ChatCompletions:
+		return fmt.Sprintf("%s/compatible-mode/v1/chat/completions", meta.BaseURL), nil
+	case relaymode.Embeddings:
+		return fmt.Sprintf("%s/compatible-mode/v1/embeddings", meta.BaseURL), nil
+	default:
+	}
+	return "", fmt.Errorf("unsupported relay mode %d for ali bailian", meta.Mode)
+}
--- a/relay/adaptor/baiduv2/constants.go
+++ b/relay/adaptor/baiduv2/constants.go
@@ -0,0 +1,30 @@
+package baiduv2
+
+// https://console.bce.baidu.com/support/?_=1692863460488&timestamp=1739074632076#/api?product=QIANFAN&project=%E5%8D%83%E5%B8%86ModelBuilder&parent=%E5%AF%B9%E8%AF%9DChat%20V2&api=v2%2Fchat%2Fcompletions&method=post
+// https://cloud.baidu.com/doc/WENXINWORKSHOP/s/Fm2vrveyu#%E6%94%AF%E6%8C%81%E6%A8%A1%E5%9E%8B%E5%88%97%E8%A1%A8
+
+var ModelList = []string{
+	"ernie-4.0-8k-latest",
+	"ernie-4.0-8k-preview",
+	"ernie-4.0-8k",
+	"ernie-4.0-turbo-8k-latest",
+	"ernie-4.0-turbo-8k-preview",
+	"ernie-4.0-turbo-8k",
+	"ernie-4.0-turbo-128k",
+	"ernie-3.5-8k-preview",
+	"ernie-3.5-8k",
+	"ernie-3.5-128k",
+	"ernie-speed-8k",
+	"ernie-speed-128k",
+	"ernie-speed-pro-128k",
+	"ernie-lite-8k",
+	"ernie-lite-pro-128k",
+	"ernie-tiny-8k",
+	"ernie-char-8k",
+	"ernie-char-fiction-8k",
+	"ernie-novel-8k",
+	"deepseek-v3",
+	"deepseek-r1",
+	"deepseek-r1-distill-qwen-32b",
+	"deepseek-r1-distill-qwen-14b",
+}
--- a/relay/adaptor/baiduv2/main.go
+++ b/relay/adaptor/baiduv2/main.go
@@ -0,0 +1,17 @@
+package baiduv2
+
+import (
+	"fmt"
+
+	"github.com/songquanpeng/one-api/relay/meta"
+	"github.com/songquanpeng/one-api/relay/relaymode"
+)
+
+func GetRequestURL(meta *meta.Meta) (string, error) {
+	switch meta.Mode {
+	case relaymode.ChatCompletions:
+		return fmt.Sprintf("%s/v2/chat/completions", meta.BaseURL), nil
+	default:
+	}
+	return "", fmt.Errorf("unsupported relay mode %d for baidu v2", meta.Mode)
+}
--- a/relay/adaptor/coze/adaptor_v3.go
+++ b/relay/adaptor/coze/adaptor_v3.go
@@ -0,0 +1,75 @@
+package coze
+
+import (
+	"errors"
+	"fmt"
+	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/relay/adaptor"
+	"github.com/songquanpeng/one-api/relay/adaptor/openai"
+	"github.com/songquanpeng/one-api/relay/meta"
+	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+)
+
+type AdaptorV3 struct {
+	meta *meta.Meta
+}
+
+func (a *AdaptorV3) Init(meta *meta.Meta) {
+	a.meta = meta
+}
+
+func (a *AdaptorV3) GetRequestURL(meta *meta.Meta) (string, error) {
+	return fmt.Sprintf("%s/v3/chat", meta.BaseURL), nil
+}
+
+func (a *AdaptorV3) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
+	adaptor.SetupCommonRequestHeader(c, req, meta)
+	req.Header.Set("Authorization", "Bearer "+meta.APIKey)
+	return nil
+}
+
+func (a *AdaptorV3) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	request.User = a.meta.Config.UserID
+	return V3ConvertRequest(*request), nil
+}
+
+func (a *AdaptorV3) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
+func (a *AdaptorV3) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
+	return adaptor.DoRequestHelper(a, c, meta, requestBody)
+}
+
+func (a *AdaptorV3) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
+	var responseText *string
+	if meta.IsStream {
+		err, responseText = V3StreamHandler(c, resp)
+	} else {
+		err, responseText = V3Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
+	}
+	if responseText != nil {
+		usage = openai.ResponseText2Usage(*responseText, meta.ActualModelName, meta.PromptTokens)
+	} else {
+		usage = &model.Usage{}
+	}
+	usage.PromptTokens = meta.PromptTokens
+	usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
+	return
+}
+
+func (a *AdaptorV3) GetModelList() []string {
+	return ModelList
+}
+
+func (a *AdaptorV3) GetChannelName() string {
+	return "CozeV3"
+}
--- a/relay/adaptor/coze/helper.go
+++ b/relay/adaptor/coze/helper.go
@@ -1,6 +1,9 @@
 package coze

-import "github.com/songquanpeng/one-api/relay/adaptor/coze/constant/event"
+import (
+	"github.com/songquanpeng/one-api/relay/adaptor/coze/constant/event"
+	"strings"
+)

 func event2StopReason(e *string) string {
 	if e == nil || *e == event.Message {
@@ -8,3 +11,16 @@ func event2StopReason(e *string) string {
 	}
 	return "stop"
 }
+
+func splitOnDoubleNewline(data []byte, atEOF bool) (advance int, token []byte, err error) {
+	if atEOF && len(data) == 0 {
+		return 0, nil, nil
+	}
+	if i := strings.Index(string(data), "\n\n"); i >= 0 {
+		return i + 1, data[0:i], nil
+	}
+	if atEOF {
+		return len(data), data, nil
+	}
+	return 0, nil, nil
+}
--- a/relay/adaptor/coze/main.go
+++ b/relay/adaptor/coze/main.go
@@ -4,19 +4,18 @@ import (
 	"bufio"
 	"encoding/json"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
-	"io"
-	"net/http"
-	"strings"
-
 	"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/common/render"
 	"github.com/songquanpeng/one-api/relay/adaptor/coze/constant/messagetype"
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+	"strings"
 )

 // https://www.coze.com/open
@@ -57,6 +56,32 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request {
 	return &cozeRequest
 }

+func V3ConvertRequest(textRequest model.GeneralOpenAIRequest) *V3Request {
+	cozeRequest := V3Request{
+		UserId: textRequest.User,
+		Stream: textRequest.Stream,
+		BotId:  strings.TrimPrefix(textRequest.Model, "bot-"),
+	}
+	if cozeRequest.UserId == "" {
+		cozeRequest.UserId = "any"
+	}
+	for i, message := range textRequest.Messages {
+		if i == len(textRequest.Messages)-1 {
+			cozeRequest.AdditionalMessages = append(cozeRequest.AdditionalMessages, Message{
+				Role:    "user",
+				Content: message.CozeV3StringContent(),
+			})
+			continue
+		}
+		cozeMessage := Message{
+			Role:    message.Role,
+			Content: message.CozeV3StringContent(),
+		}
+		cozeRequest.AdditionalMessages = append(cozeRequest.AdditionalMessages, cozeMessage)
+	}
+	return &cozeRequest
+}
+
 func StreamResponseCoze2OpenAI(cozeResponse *StreamResponse) (*openai.ChatCompletionsStreamResponse, *Response) {
 	var response *Response
 	var stopReason string
@@ -80,6 +105,28 @@ func StreamResponseCoze2OpenAI(cozeResponse *StreamResponse) (*openai.ChatComple
 	return &openaiResponse, response
 }

+func V3StreamResponseCoze2OpenAI(cozeResponse *V3StreamResponse) (*openai.ChatCompletionsStreamResponse, *Response) {
+	var response *Response
+	var choice openai.ChatCompletionsStreamResponseChoice
+
+	choice.Delta.Role = cozeResponse.Role
+	choice.Delta.Content = cozeResponse.Content
+
+	var openaiResponse openai.ChatCompletionsStreamResponse
+	openaiResponse.Object = "chat.completion.chunk"
+	openaiResponse.Choices = []openai.ChatCompletionsStreamResponseChoice{choice}
+	openaiResponse.Id = cozeResponse.ConversationId
+
+	if cozeResponse.Usage.TokenCount > 0 {
+		openaiResponse.Usage = &model.Usage{
+			PromptTokens:     cozeResponse.Usage.InputCount,
+			CompletionTokens: cozeResponse.Usage.OutputCount,
+			TotalTokens:      cozeResponse.Usage.TokenCount,
+		}
+	}
+	return &openaiResponse, response
+}
+
 func ResponseCoze2OpenAI(cozeResponse *Response) *openai.TextResponse {
 	var responseText string
 	for _, message := range cozeResponse.Messages {
@@ -107,6 +154,26 @@ func ResponseCoze2OpenAI(cozeResponse *Response) *openai.TextResponse {
 	return &fullTextResponse
 }

+func V3ResponseCoze2OpenAI(cozeResponse *V3Response) *openai.TextResponse {
+	choice := openai.TextResponseChoice{
+		Index: 0,
+		Message: model.Message{
+			Role:    "assistant",
+			Content: cozeResponse.Data.Content,
+			Name:    nil,
+		},
+		FinishReason: "stop",
+	}
+	fullTextResponse := openai.TextResponse{
+		Id:      fmt.Sprintf("chatcmpl-%s", cozeResponse.Data.ConversationId),
+		Model:   "coze-bot",
+		Object:  "chat.completion",
+		Created: helper.GetTimestamp(),
+		Choices: []openai.TextResponseChoice{choice},
+	}
+	return &fullTextResponse
+}
+
 func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *string) {
 	var responseText string
 	createdTime := helper.GetTimestamp()
@@ -162,6 +229,63 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 	return nil, &responseText
 }

+func V3StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *string) {
+	var responseText string
+	createdTime := helper.GetTimestamp()
+	scanner := bufio.NewScanner(resp.Body)
+	scanner.Split(splitOnDoubleNewline)
+	common.SetEventStreamHeaders(c)
+	var modelName string
+	for scanner.Scan() {
+		part := scanner.Text()
+		part = strings.TrimPrefix(part, "\n")
+		parts := strings.Split(part, "\n")
+		if len(parts) != 2 {
+			continue
+		}
+		if !strings.HasPrefix(parts[0], "event:") || !strings.HasPrefix(parts[1], "data:") {
+			continue
+		}
+		event, data := strings.TrimSpace(parts[0][6:]), strings.TrimSpace(parts[1][5:])
+		if event == "conversation.message.delta" || event == "conversation.chat.completed" {
+			data = strings.TrimSuffix(data, "\r")
+			var cozeResponse V3StreamResponse
+			err := json.Unmarshal([]byte(data), &cozeResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				continue
+			}
+
+			response, _ := V3StreamResponseCoze2OpenAI(&cozeResponse)
+			if response == nil {
+				continue
+			}
+
+			for _, choice := range response.Choices {
+				responseText += conv.AsString(choice.Delta.Content)
+			}
+			response.Model = modelName
+			response.Created = createdTime
+
+			err = render.ObjectData(c, response)
+			if err != nil {
+				logger.SysError(err.Error())
+			}
+		}
+	}
+	if err := scanner.Err(); err != nil {
+		logger.SysError("error reading stream: " + err.Error())
+	}
+
+	render.Done(c)
+	err := resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+
+	return nil, &responseText
+}
+
 func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName string) (*model.ErrorWithStatusCode, *string) {
 	responseBody, err := io.ReadAll(resp.Body)
 	if err != nil {
@@ -200,3 +324,42 @@ func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName st
 	}
 	return nil, &responseText
 }
+
+func V3Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName string) (*model.ErrorWithStatusCode, *string) {
+	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
+	}
+	var cozeResponse V3Response
+	err = json.Unmarshal(responseBody, &cozeResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	if cozeResponse.Code != 0 {
+		return &model.ErrorWithStatusCode{
+			Error: model.Error{
+				Message: cozeResponse.Msg,
+				Code:    cozeResponse.Code,
+			},
+			StatusCode: resp.StatusCode,
+		}, nil
+	}
+	fullTextResponse := V3ResponseCoze2OpenAI(&cozeResponse)
+	fullTextResponse.Model = modelName
+	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)
+	var responseText string
+	if len(fullTextResponse.Choices) > 0 {
+		responseText = fullTextResponse.Choices[0].Message.StringContent()
+	}
+	return nil, &responseText
+}
--- a/relay/adaptor/coze/model.go
+++ b/relay/adaptor/coze/model.go
@@ -2,9 +2,9 @@ package coze

 type Message struct {
 	Role        string `json:"role"`
-	Type        string `json:"type"`
+	Type        string `json:"type,omitempty"`
 	Content     string `json:"content"`
-	ContentType string `json:"content_type"`
+	ContentType string `json:"content_type,omitempty"`
 }

 type ErrorInformation struct {
@@ -36,3 +36,52 @@ type StreamResponse struct {
 	ConversationId   string            `json:"conversation_id,omitempty"`
 	ErrorInformation *ErrorInformation `json:"error_information,omitempty"`
 }
+
+type V3StreamResponse struct {
+	Id             string `json:"id"`
+	ConversationId string `json:"conversation_id"`
+	BotId          string `json:"bot_id"`
+	Role           string `json:"role"`
+	Type           string `json:"type"`
+	Content        string `json:"content"`
+	ContentType    string `json:"content_type"`
+	ChatId         string `json:"chat_id"`
+	CreatedAt      int    `json:"created_at"`
+	CompletedAt    int    `json:"completed_at"`
+	LastError      struct {
+		Code int    `json:"code"`
+		Msg  string `json:"msg"`
+	} `json:"last_error"`
+	Status string `json:"status"`
+	Usage  struct {
+		TokenCount  int `json:"token_count"`
+		OutputCount int `json:"output_count"`
+		InputCount  int `json:"input_count"`
+	} `json:"usage"`
+	SectionId string `json:"section_id"`
+}
+
+type V3Response struct {
+	Data struct {
+		Id             string `json:"id"`
+		ConversationId string `json:"conversation_id"`
+		BotId          string `json:"bot_id"`
+		Content        string `json:"content"`
+		ContentType    string `json:"content_type"`
+		CreatedAt      int    `json:"created_at"`
+		LastError      struct {
+			Code int    `json:"code"`
+			Msg  string `json:"msg"`
+		} `json:"last_error"`
+		Status string `json:"status"`
+	} `json:"data"`
+	Code int    `json:"code"`
+	Msg  string `json:"msg"`
+}
+
+type V3Request struct {
+	BotId              string    `json:"bot_id"`
+	UserId             string    `json:"user_id"`
+	AdditionalMessages []Message `json:"additional_messages"`
+	Stream             bool      `json:"stream"`
+}
--- a/relay/adaptor/gemini/adaptor.go
+++ b/relay/adaptor/gemini/adaptor.go
@@ -5,9 +5,10 @@ import (
 	"fmt"
 	"io"
 	"net/http"
+	"strings"

 	"github.com/gin-gonic/gin"
-
+	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/common/helper"
 	channelhelper "github.com/songquanpeng/one-api/relay/adaptor"
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
@@ -20,17 +21,12 @@ type Adaptor struct {
 }

 func (a *Adaptor) Init(meta *meta.Meta) {
-
 }

 func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
-	var defaultVersion string
-	switch meta.ActualModelName {
-	case "gemini-2.0-flash-exp",
-		"gemini-2.0-flash-thinking-exp",
-		"gemini-2.0-flash-thinking-exp-01-21":
-		defaultVersion = "v1beta"
-	default:
+	defaultVersion := config.GeminiVersion
+	if strings.Contains(meta.ActualModelName, "gemini-2") ||
+		strings.Contains(meta.ActualModelName, "gemini-1.5") {
 		defaultVersion = "v1beta"
 	}

--- a/relay/adaptor/gemini/constants.go
+++ b/relay/adaptor/gemini/constants.go
@@ -1,11 +1,35 @@
 package gemini

+import (
+	"github.com/songquanpeng/one-api/relay/adaptor/geminiv2"
+)
+
 // https://ai.google.dev/models/gemini

-var ModelList = []string{
-	"gemini-pro", "gemini-1.0-pro",
-	"gemini-1.5-flash", "gemini-1.5-pro",
-	"text-embedding-004", "aqa",
-	"gemini-2.0-flash-exp",
-	"gemini-2.0-flash-thinking-exp", "gemini-2.0-flash-thinking-exp-01-21",
+var ModelList = geminiv2.ModelList
+
+// ModelsSupportSystemInstruction is the list of models that support system instruction.
+//
+// https://cloud.google.com/vertex-ai/generative-ai/docs/learn/prompts/system-instructions
+var ModelsSupportSystemInstruction = []string{
+	// "gemini-1.0-pro-002",
+	// "gemini-1.5-flash", "gemini-1.5-flash-001", "gemini-1.5-flash-002",
+	// "gemini-1.5-flash-8b",
+	// "gemini-1.5-pro", "gemini-1.5-pro-001", "gemini-1.5-pro-002",
+	// "gemini-1.5-pro-experimental",
+	"gemini-2.0-flash", "gemini-2.0-flash-exp",
+	"gemini-2.0-flash-thinking-exp-01-21",
+}
+
+// IsModelSupportSystemInstruction check if the model support system instruction.
+//
+// Because the main version of Go is 1.20, slice.Contains cannot be used
+func IsModelSupportSystemInstruction(model string) bool {
+	for _, m := range ModelsSupportSystemInstruction {
+		if m == model {
+			return true
+		}
+	}
+
+	return false
 }
--- a/relay/adaptor/gemini/main.go
+++ b/relay/adaptor/gemini/main.go
@@ -12,6 +12,7 @@ import (

 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/common/file"
 	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/image"
 	"github.com/songquanpeng/one-api/common/logger"
@@ -66,6 +67,23 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *ChatRequest {
 			MaxOutputTokens: textRequest.MaxTokens,
 		},
 	}
+
+	if textRequest.ReasoningEffort != nil {
+		var thinkBudget int
+		switch *textRequest.ReasoningEffort {
+		case "low":
+			thinkBudget = 1000
+		case "medium":
+			thinkBudget = 8000
+		case "high":
+			thinkBudget = 24000
+		}
+		geminiRequest.GenerationConfig.ThinkingConfig = &ThinkingConfig{
+			ThinkingBudget:  thinkBudget,
+			IncludeThoughts: true,
+		}
+	}
+
 	if textRequest.ResponseFormat != nil {
 		if mimeType, ok := mimeTypeMap[textRequest.ResponseFormat.Type]; ok {
 			geminiRequest.GenerationConfig.ResponseMimeType = mimeType
@@ -76,22 +94,13 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *ChatRequest {
 		}
 	}
 	if textRequest.Tools != nil {
-		functions := make([]model.Function, 0, len(textRequest.Tools))
-		for _, tool := range textRequest.Tools {
-			functions = append(functions, tool.Function)
-		}
-		geminiRequest.Tools = []ChatTools{
-			{
-				FunctionDeclarations: functions,
-			},
-		}
-	} else if textRequest.Functions != nil {
-		geminiRequest.Tools = []ChatTools{
-			{
-				FunctionDeclarations: textRequest.Functions,
-			},
-		}
+		geminiRequest.Tools = textRequest.Tools
 	}
+
+	if textRequest.Functions != nil {
+		geminiRequest.Tools = textRequest.Functions
+	}
+
 	shouldAddDummyModelMessage := false
 	for _, message := range textRequest.Messages {
 		content := ChatContent{
@@ -110,6 +119,14 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *ChatRequest {
 				parts = append(parts, Part{
 					Text: part.Text,
 				})
+			} else if part.Type == model.ContentTypeInputFile {
+				mimeType, data, _ := file.GetFileFromUrl(part.File.FileData)
+				parts = append(parts, Part{
+					InlineData: &InlineData{
+						MimeType: mimeType,
+						Data:     data,
+					},
+				})
 			} else if part.Type == model.ContentTypeImageURL {
 				imageNum += 1
 				if imageNum > VisionMaxImageNum {
@@ -132,9 +149,16 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *ChatRequest {
 		}
 		// Converting system prompt to prompt from user for the same reason
 		if content.Role == "system" {
-			content.Role = "user"
 			shouldAddDummyModelMessage = true
+			if IsModelSupportSystemInstruction(textRequest.Model) {
+				geminiRequest.SystemInstruction = &content
+				geminiRequest.SystemInstruction.Role = ""
+				continue
+			} else {
+				content.Role = "user"
+			}
 		}
+
 		geminiRequest.Contents = append(geminiRequest.Contents, content)

 		// If a system message is the last message, we need to add a dummy model message to make gemini happy
@@ -192,6 +216,21 @@ func (g *ChatResponse) GetResponseText() string {
 	return ""
 }

+func (g *ChatResponse) GetResponseTextAndThought() (content string, thought string) {
+	if g == nil {
+		return
+	}
+	if len(g.Candidates) > 0 && len(g.Candidates[0].Content.Parts) > 0 {
+		contentPart := g.Candidates[0].Content.Parts[0]
+		if contentPart.Thought {
+			thought = contentPart.Text
+			return
+		}
+		content = contentPart.Text
+	}
+	return
+}
+
 type ChatCandidate struct {
 	Content       ChatContent        `json:"content"`
 	FinishReason  string             `json:"finishReason"`
@@ -256,7 +295,11 @@ func responseGeminiChat2OpenAI(response *ChatResponse) *openai.TextResponse {
 					if i > 0 {
 						builder.WriteString("\n")
 					}
-					builder.WriteString(part.Text)
+					if part.Thought {
+						builder.WriteString(fmt.Sprintf("<think>%s</think>\n", part.Text))
+					} else {
+						builder.WriteString(part.Text)
+					}
 				}
 				choice.Message.Content = builder.String()
 			}
@@ -271,7 +314,7 @@ func responseGeminiChat2OpenAI(response *ChatResponse) *openai.TextResponse {

 func streamResponseGeminiChat2OpenAI(geminiResponse *ChatResponse) *openai.ChatCompletionsStreamResponse {
 	var choice openai.ChatCompletionsStreamResponseChoice
-	choice.Delta.Content = geminiResponse.GetResponseText()
+	choice.Delta.Content, choice.Delta.ReasoningContent = geminiResponse.GetResponseTextAndThought()
 	//choice.FinishReason = &constant.StopFinishReason
 	var response openai.ChatCompletionsStreamResponse
 	response.Id = fmt.Sprintf("chatcmpl-%s", random.GetUUID())
--- a/relay/adaptor/gemini/model.go
+++ b/relay/adaptor/gemini/model.go
@@ -1,10 +1,11 @@
 package gemini

 type ChatRequest struct {
-	Contents         []ChatContent        `json:"contents"`
-	SafetySettings   []ChatSafetySettings `json:"safety_settings,omitempty"`
-	GenerationConfig ChatGenerationConfig `json:"generation_config,omitempty"`
-	Tools            []ChatTools          `json:"tools,omitempty"`
+	Contents          []ChatContent        `json:"contents"`
+	SafetySettings    []ChatSafetySettings `json:"safety_settings,omitempty"`
+	GenerationConfig  ChatGenerationConfig `json:"generation_config,omitempty"`
+	Tools             interface{}          `json:"tools,omitempty"`
+	SystemInstruction *ChatContent         `json:"system_instruction,omitempty"`
 }

 type EmbeddingRequest struct {
@@ -39,6 +40,11 @@ type InlineData struct {
 	Data     string `json:"data"`
 }

+type FileData struct {
+	MimeType string `json:"mime_type"`
+	FileUri  string `json:"file_uri"`
+}
+
 type FunctionCall struct {
 	FunctionName string `json:"name"`
 	Arguments    any    `json:"args"`
@@ -48,6 +54,8 @@ type Part struct {
 	Text         string        `json:"text,omitempty"`
 	InlineData   *InlineData   `json:"inlineData,omitempty"`
 	FunctionCall *FunctionCall `json:"functionCall,omitempty"`
+	Thought      bool          `json:"thought,omitempty"`
+	FileData     *FileData     `json:"fileData,omitempty"`
 }

 type ChatContent struct {
@@ -65,12 +73,18 @@ type ChatTools struct {
 }

 type ChatGenerationConfig struct {
-	ResponseMimeType string   `json:"responseMimeType,omitempty"`
-	ResponseSchema   any      `json:"responseSchema,omitempty"`
-	Temperature      *float64 `json:"temperature,omitempty"`
-	TopP             *float64 `json:"topP,omitempty"`
-	TopK             float64  `json:"topK,omitempty"`
-	MaxOutputTokens  int      `json:"maxOutputTokens,omitempty"`
-	CandidateCount   int      `json:"candidateCount,omitempty"`
-	StopSequences    []string `json:"stopSequences,omitempty"`
+	ResponseMimeType string          `json:"responseMimeType,omitempty"`
+	ResponseSchema   any             `json:"responseSchema,omitempty"`
+	Temperature      *float64        `json:"temperature,omitempty"`
+	TopP             *float64        `json:"topP,omitempty"`
+	TopK             float64         `json:"topK,omitempty"`
+	MaxOutputTokens  int             `json:"maxOutputTokens,omitempty"`
+	CandidateCount   int             `json:"candidateCount,omitempty"`
+	StopSequences    []string        `json:"stopSequences,omitempty"`
+	ThinkingConfig   *ThinkingConfig `json:"thinkingConfig,omitempty"`
+}
+
+type ThinkingConfig struct {
+	ThinkingBudget  int  `json:"thinkingBudget"`
+	IncludeThoughts bool `json:"includeThoughts"`
 }
--- a/relay/adaptor/geminiv2/constants.go
+++ b/relay/adaptor/geminiv2/constants.go
@@ -0,0 +1,15 @@
+package geminiv2
+
+// https://ai.google.dev/models/gemini
+
+var ModelList = []string{
+	"gemini-pro", "gemini-1.0-pro",
+	// "gemma-2-2b-it", "gemma-2-9b-it", "gemma-2-27b-it",
+	"gemini-1.5-flash", "gemini-1.5-flash-8b",
+	"gemini-1.5-pro", "gemini-1.5-pro-experimental",
+	"text-embedding-004", "aqa",
+	"gemini-2.0-flash", "gemini-2.0-flash-exp",
+	"gemini-2.0-flash-lite-preview-02-05",
+	"gemini-2.0-flash-thinking-exp-01-21",
+	"gemini-2.0-pro-exp-02-05",
+}
--- a/relay/adaptor/geminiv2/main.go
+++ b/relay/adaptor/geminiv2/main.go
@@ -0,0 +1,14 @@
+package geminiv2
+
+import (
+	"fmt"
+	"strings"
+
+	"github.com/songquanpeng/one-api/relay/meta"
+)
+
+func GetRequestURL(meta *meta.Meta) (string, error) {
+	baseURL := strings.TrimSuffix(meta.BaseURL, "/")
+	requestPath := strings.TrimPrefix(meta.RequestURLPath, "/v1")
+	return fmt.Sprintf("%s%s", baseURL, requestPath), nil
+}
--- a/relay/adaptor/groq/constants.go
+++ b/relay/adaptor/groq/constants.go
@@ -3,7 +3,6 @@ package groq
 // https://console.groq.com/docs/models

 var ModelList = []string{
-	"gemma-7b-it",
 	"gemma2-9b-it",
 	"llama-3.1-70b-versatile",
 	"llama-3.1-8b-instant",
@@ -23,4 +22,6 @@ var ModelList = []string{
 	"distil-whisper-large-v3-en",
 	"whisper-large-v3",
 	"whisper-large-v3-turbo",
+	"deepseek-r1-distill-llama-70b-specdec",
+	"deepseek-r1-distill-llama-70b",
 }
--- a/relay/adaptor/minimax/constants.go
+++ b/relay/adaptor/minimax/constants.go
@@ -8,4 +8,6 @@ var ModelList = []string{
 	"abab6-chat",
 	"abab5.5-chat",
 	"abab5.5s-chat",
+	"MiniMax-VL-01",
+	"MiniMax-Text-01",
 }
--- a/relay/adaptor/openai/adaptor.go
+++ b/relay/adaptor/openai/adaptor.go
@@ -8,8 +8,12 @@ import (
 	"strings"

 	"github.com/gin-gonic/gin"
+
 	"github.com/songquanpeng/one-api/relay/adaptor"
+	"github.com/songquanpeng/one-api/relay/adaptor/alibailian"
+	"github.com/songquanpeng/one-api/relay/adaptor/baiduv2"
 	"github.com/songquanpeng/one-api/relay/adaptor/doubao"
+	"github.com/songquanpeng/one-api/relay/adaptor/geminiv2"
 	"github.com/songquanpeng/one-api/relay/adaptor/minimax"
 	"github.com/songquanpeng/one-api/relay/adaptor/novita"
 	"github.com/songquanpeng/one-api/relay/channeltype"
@@ -41,7 +45,6 @@ func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
 		requestURL = fmt.Sprintf("%s?api-version=%s", requestURL, meta.Config.APIVersion)
 		task := strings.TrimPrefix(requestURL, "/v1/")
 		model_ := meta.ActualModelName
-		model_ = strings.Replace(model_, ".", "", -1)
 		//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)
@@ -52,6 +55,12 @@ func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
 		return doubao.GetRequestURL(meta)
 	case channeltype.Novita:
 		return novita.GetRequestURL(meta)
+	case channeltype.BaiduV2:
+		return baiduv2.GetRequestURL(meta)
+	case channeltype.AliBailian:
+		return alibailian.GetRequestURL(meta)
+	case channeltype.GeminiOpenAICompatible:
+		return geminiv2.GetRequestURL(meta)
 	default:
 		return GetFullRequestURL(meta.BaseURL, meta.RequestURLPath, meta.ChannelType), nil
 	}
--- a/relay/adaptor/openai/compatible.go
+++ b/relay/adaptor/openai/compatible.go
@@ -2,19 +2,24 @@ package openai

 import (
 	"github.com/songquanpeng/one-api/relay/adaptor/ai360"
+	"github.com/songquanpeng/one-api/relay/adaptor/alibailian"
 	"github.com/songquanpeng/one-api/relay/adaptor/baichuan"
+	"github.com/songquanpeng/one-api/relay/adaptor/baiduv2"
 	"github.com/songquanpeng/one-api/relay/adaptor/deepseek"
 	"github.com/songquanpeng/one-api/relay/adaptor/doubao"
+	"github.com/songquanpeng/one-api/relay/adaptor/geminiv2"
 	"github.com/songquanpeng/one-api/relay/adaptor/groq"
 	"github.com/songquanpeng/one-api/relay/adaptor/lingyiwanwu"
 	"github.com/songquanpeng/one-api/relay/adaptor/minimax"
 	"github.com/songquanpeng/one-api/relay/adaptor/mistral"
 	"github.com/songquanpeng/one-api/relay/adaptor/moonshot"
 	"github.com/songquanpeng/one-api/relay/adaptor/novita"
+	"github.com/songquanpeng/one-api/relay/adaptor/openrouter"
 	"github.com/songquanpeng/one-api/relay/adaptor/siliconflow"
 	"github.com/songquanpeng/one-api/relay/adaptor/stepfun"
 	"github.com/songquanpeng/one-api/relay/adaptor/togetherai"
 	"github.com/songquanpeng/one-api/relay/adaptor/xai"
+	"github.com/songquanpeng/one-api/relay/adaptor/xunfeiv2"
 	"github.com/songquanpeng/one-api/relay/channeltype"
 )

@@ -34,6 +39,8 @@ var CompatibleChannels = []int{
 	channeltype.Novita,
 	channeltype.SiliconFlow,
 	channeltype.XAI,
+	channeltype.BaiduV2,
+	channeltype.XunfeiV2,
 }

 func GetCompatibleChannelMeta(channelType int) (string, []string) {
@@ -68,6 +75,16 @@ func GetCompatibleChannelMeta(channelType int) (string, []string) {
 		return "siliconflow", siliconflow.ModelList
 	case channeltype.XAI:
 		return "xai", xai.ModelList
+	case channeltype.BaiduV2:
+		return "baiduv2", baiduv2.ModelList
+	case channeltype.XunfeiV2:
+		return "xunfeiv2", xunfeiv2.ModelList
+	case channeltype.OpenRouter:
+		return "openrouter", openrouter.ModelList
+	case channeltype.AliBailian:
+		return "alibailian", alibailian.ModelList
+	case channeltype.GeminiOpenAICompatible:
+		return "geminiv2", geminiv2.ModelList
 	default:
 		return "openai", ModelList
 	}
--- a/relay/adaptor/openai/constants.go
+++ b/relay/adaptor/openai/constants.go
@@ -4,7 +4,7 @@ var ModelList = []string{
 	"gpt-3.5-turbo", "gpt-3.5-turbo-0301", "gpt-3.5-turbo-0613", "gpt-3.5-turbo-1106", "gpt-3.5-turbo-0125",
 	"gpt-3.5-turbo-16k", "gpt-3.5-turbo-16k-0613",
 	"gpt-3.5-turbo-instruct",
-	"gpt-4", "gpt-4-0314", "gpt-4-0613", "gpt-4-1106-preview", "gpt-4-0125-preview",
+	"gpt-4", "gpt-4.1", "gpt-4-0314", "gpt-4-0613", "gpt-4-1106-preview", "gpt-4-0125-preview",
 	"gpt-4-32k", "gpt-4-32k-0314", "gpt-4-32k-0613",
 	"gpt-4-turbo-preview", "gpt-4-turbo", "gpt-4-turbo-2024-04-09",
 	"gpt-4o", "gpt-4o-2024-05-13",
--- a/relay/adaptor/openai/helper.go
+++ b/relay/adaptor/openai/helper.go
@@ -17,6 +17,9 @@ func ResponseText2Usage(responseText string, modelName string, promptTokens int)
 }

 func GetFullRequestURL(baseURL string, requestURL string, channelType int) string {
+	if channelType == channeltype.OpenAICompatible {
+		return fmt.Sprintf("%s%s", strings.TrimSuffix(baseURL, "/"), strings.TrimPrefix(requestURL, "/v1"))
+	}
 	fullRequestURL := fmt.Sprintf("%s%s", baseURL, requestURL)

 	if strings.HasPrefix(baseURL, "https://gateway.ai.cloudflare.com") {
--- a/relay/adaptor/openrouter/constants.go
+++ b/relay/adaptor/openrouter/constants.go
@@ -0,0 +1,235 @@
+package openrouter
+
+var ModelList = []string{
+	"01-ai/yi-large",
+	"aetherwiing/mn-starcannon-12b",
+	"ai21/jamba-1-5-large",
+	"ai21/jamba-1-5-mini",
+	"ai21/jamba-instruct",
+	"aion-labs/aion-1.0",
+	"aion-labs/aion-1.0-mini",
+	"aion-labs/aion-rp-llama-3.1-8b",
+	"allenai/llama-3.1-tulu-3-405b",
+	"alpindale/goliath-120b",
+	"alpindale/magnum-72b",
+	"amazon/nova-lite-v1",
+	"amazon/nova-micro-v1",
+	"amazon/nova-pro-v1",
+	"anthracite-org/magnum-v2-72b",
+	"anthracite-org/magnum-v4-72b",
+	"anthropic/claude-2",
+	"anthropic/claude-2.0",
+	"anthropic/claude-2.0:beta",
+	"anthropic/claude-2.1",
+	"anthropic/claude-2.1:beta",
+	"anthropic/claude-2:beta",
+	"anthropic/claude-3-haiku",
+	"anthropic/claude-3-haiku:beta",
+	"anthropic/claude-3-opus",
+	"anthropic/claude-3-opus:beta",
+	"anthropic/claude-3-sonnet",
+	"anthropic/claude-3-sonnet:beta",
+	"anthropic/claude-3.5-haiku",
+	"anthropic/claude-3.5-haiku-20241022",
+	"anthropic/claude-3.5-haiku-20241022:beta",
+	"anthropic/claude-3.5-haiku:beta",
+	"anthropic/claude-3.5-sonnet",
+	"anthropic/claude-3.5-sonnet-20240620",
+	"anthropic/claude-3.5-sonnet-20240620:beta",
+	"anthropic/claude-3.5-sonnet:beta",
+	"cognitivecomputations/dolphin-mixtral-8x22b",
+	"cognitivecomputations/dolphin-mixtral-8x7b",
+	"cohere/command",
+	"cohere/command-r",
+	"cohere/command-r-03-2024",
+	"cohere/command-r-08-2024",
+	"cohere/command-r-plus",
+	"cohere/command-r-plus-04-2024",
+	"cohere/command-r-plus-08-2024",
+	"cohere/command-r7b-12-2024",
+	"databricks/dbrx-instruct",
+	"deepseek/deepseek-chat",
+	"deepseek/deepseek-chat-v2.5",
+	"deepseek/deepseek-chat:free",
+	"deepseek/deepseek-r1",
+	"deepseek/deepseek-r1-distill-llama-70b",
+	"deepseek/deepseek-r1-distill-llama-70b:free",
+	"deepseek/deepseek-r1-distill-llama-8b",
+	"deepseek/deepseek-r1-distill-qwen-1.5b",
+	"deepseek/deepseek-r1-distill-qwen-14b",
+	"deepseek/deepseek-r1-distill-qwen-32b",
+	"deepseek/deepseek-r1:free",
+	"eva-unit-01/eva-llama-3.33-70b",
+	"eva-unit-01/eva-qwen-2.5-32b",
+	"eva-unit-01/eva-qwen-2.5-72b",
+	"google/gemini-2.0-flash-001",
+	"google/gemini-2.0-flash-exp:free",
+	"google/gemini-2.0-flash-lite-preview-02-05:free",
+	"google/gemini-2.0-flash-thinking-exp-1219:free",
+	"google/gemini-2.0-flash-thinking-exp:free",
+	"google/gemini-2.0-pro-exp-02-05:free",
+	"google/gemini-exp-1206:free",
+	"google/gemini-flash-1.5",
+	"google/gemini-flash-1.5-8b",
+	"google/gemini-flash-1.5-8b-exp",
+	"google/gemini-pro",
+	"google/gemini-pro-1.5",
+	"google/gemini-pro-vision",
+	"google/gemma-2-27b-it",
+	"google/gemma-2-9b-it",
+	"google/gemma-2-9b-it:free",
+	"google/gemma-7b-it",
+	"google/learnlm-1.5-pro-experimental:free",
+	"google/palm-2-chat-bison",
+	"google/palm-2-chat-bison-32k",
+	"google/palm-2-codechat-bison",
+	"google/palm-2-codechat-bison-32k",
+	"gryphe/mythomax-l2-13b",
+	"gryphe/mythomax-l2-13b:free",
+	"huggingfaceh4/zephyr-7b-beta:free",
+	"infermatic/mn-inferor-12b",
+	"inflection/inflection-3-pi",
+	"inflection/inflection-3-productivity",
+	"jondurbin/airoboros-l2-70b",
+	"liquid/lfm-3b",
+	"liquid/lfm-40b",
+	"liquid/lfm-7b",
+	"mancer/weaver",
+	"meta-llama/llama-2-13b-chat",
+	"meta-llama/llama-2-70b-chat",
+	"meta-llama/llama-3-70b-instruct",
+	"meta-llama/llama-3-8b-instruct",
+	"meta-llama/llama-3-8b-instruct:free",
+	"meta-llama/llama-3.1-405b",
+	"meta-llama/llama-3.1-405b-instruct",
+	"meta-llama/llama-3.1-70b-instruct",
+	"meta-llama/llama-3.1-8b-instruct",
+	"meta-llama/llama-3.2-11b-vision-instruct",
+	"meta-llama/llama-3.2-11b-vision-instruct:free",
+	"meta-llama/llama-3.2-1b-instruct",
+	"meta-llama/llama-3.2-3b-instruct",
+	"meta-llama/llama-3.2-90b-vision-instruct",
+	"meta-llama/llama-3.3-70b-instruct",
+	"meta-llama/llama-3.3-70b-instruct:free",
+	"meta-llama/llama-guard-2-8b",
+	"microsoft/phi-3-medium-128k-instruct",
+	"microsoft/phi-3-medium-128k-instruct:free",
+	"microsoft/phi-3-mini-128k-instruct",
+	"microsoft/phi-3-mini-128k-instruct:free",
+	"microsoft/phi-3.5-mini-128k-instruct",
+	"microsoft/phi-4",
+	"microsoft/wizardlm-2-7b",
+	"microsoft/wizardlm-2-8x22b",
+	"minimax/minimax-01",
+	"mistralai/codestral-2501",
+	"mistralai/codestral-mamba",
+	"mistralai/ministral-3b",
+	"mistralai/ministral-8b",
+	"mistralai/mistral-7b-instruct",
+	"mistralai/mistral-7b-instruct-v0.1",
+	"mistralai/mistral-7b-instruct-v0.3",
+	"mistralai/mistral-7b-instruct:free",
+	"mistralai/mistral-large",
+	"mistralai/mistral-large-2407",
+	"mistralai/mistral-large-2411",
+	"mistralai/mistral-medium",
+	"mistralai/mistral-nemo",
+	"mistralai/mistral-nemo:free",
+	"mistralai/mistral-small",
+	"mistralai/mistral-small-24b-instruct-2501",
+	"mistralai/mistral-small-24b-instruct-2501:free",
+	"mistralai/mistral-tiny",
+	"mistralai/mixtral-8x22b-instruct",
+	"mistralai/mixtral-8x7b",
+	"mistralai/mixtral-8x7b-instruct",
+	"mistralai/pixtral-12b",
+	"mistralai/pixtral-large-2411",
+	"neversleep/llama-3-lumimaid-70b",
+	"neversleep/llama-3-lumimaid-8b",
+	"neversleep/llama-3-lumimaid-8b:extended",
+	"neversleep/llama-3.1-lumimaid-70b",
+	"neversleep/llama-3.1-lumimaid-8b",
+	"neversleep/noromaid-20b",
+	"nothingiisreal/mn-celeste-12b",
+	"nousresearch/hermes-2-pro-llama-3-8b",
+	"nousresearch/hermes-3-llama-3.1-405b",
+	"nousresearch/hermes-3-llama-3.1-70b",
+	"nousresearch/nous-hermes-2-mixtral-8x7b-dpo",
+	"nousresearch/nous-hermes-llama2-13b",
+	"nvidia/llama-3.1-nemotron-70b-instruct",
+	"nvidia/llama-3.1-nemotron-70b-instruct:free",
+	"openai/chatgpt-4o-latest",
+	"openai/gpt-3.5-turbo",
+	"openai/gpt-3.5-turbo-0125",
+	"openai/gpt-3.5-turbo-0613",
+	"openai/gpt-3.5-turbo-1106",
+	"openai/gpt-3.5-turbo-16k",
+	"openai/gpt-3.5-turbo-instruct",
+	"openai/gpt-4",
+	"openai/gpt-4-0314",
+	"openai/gpt-4-1106-preview",
+	"openai/gpt-4-32k",
+	"openai/gpt-4-32k-0314",
+	"openai/gpt-4-turbo",
+	"openai/gpt-4-turbo-preview",
+	"openai/gpt-4o",
+	"openai/gpt-4o-2024-05-13",
+	"openai/gpt-4o-2024-08-06",
+	"openai/gpt-4o-2024-11-20",
+	"openai/gpt-4o-mini",
+	"openai/gpt-4o-mini-2024-07-18",
+	"openai/gpt-4o:extended",
+	"openai/o1",
+	"openai/o1-mini",
+	"openai/o1-mini-2024-09-12",
+	"openai/o1-preview",
+	"openai/o1-preview-2024-09-12",
+	"openai/o3-mini",
+	"openai/o3-mini-high",
+	"openchat/openchat-7b",
+	"openchat/openchat-7b:free",
+	"openrouter/auto",
+	"perplexity/llama-3.1-sonar-huge-128k-online",
+	"perplexity/llama-3.1-sonar-large-128k-chat",
+	"perplexity/llama-3.1-sonar-large-128k-online",
+	"perplexity/llama-3.1-sonar-small-128k-chat",
+	"perplexity/llama-3.1-sonar-small-128k-online",
+	"perplexity/sonar",
+	"perplexity/sonar-reasoning",
+	"pygmalionai/mythalion-13b",
+	"qwen/qvq-72b-preview",
+	"qwen/qwen-2-72b-instruct",
+	"qwen/qwen-2-7b-instruct",
+	"qwen/qwen-2-7b-instruct:free",
+	"qwen/qwen-2-vl-72b-instruct",
+	"qwen/qwen-2-vl-7b-instruct",
+	"qwen/qwen-2.5-72b-instruct",
+	"qwen/qwen-2.5-7b-instruct",
+	"qwen/qwen-2.5-coder-32b-instruct",
+	"qwen/qwen-max",
+	"qwen/qwen-plus",
+	"qwen/qwen-turbo",
+	"qwen/qwen-vl-plus:free",
+	"qwen/qwen2.5-vl-72b-instruct:free",
+	"qwen/qwq-32b-preview",
+	"raifle/sorcererlm-8x22b",
+	"sao10k/fimbulvetr-11b-v2",
+	"sao10k/l3-euryale-70b",
+	"sao10k/l3-lunaris-8b",
+	"sao10k/l3.1-70b-hanami-x1",
+	"sao10k/l3.1-euryale-70b",
+	"sao10k/l3.3-euryale-70b",
+	"sophosympatheia/midnight-rose-70b",
+	"sophosympatheia/rogue-rose-103b-v0.2:free",
+	"teknium/openhermes-2.5-mistral-7b",
+	"thedrummer/rocinante-12b",
+	"thedrummer/unslopnemo-12b",
+	"undi95/remm-slerp-l2-13b",
+	"undi95/toppy-m-7b",
+	"undi95/toppy-m-7b:free",
+	"x-ai/grok-2-1212",
+	"x-ai/grok-2-vision-1212",
+	"x-ai/grok-beta",
+	"x-ai/grok-vision-beta",
+	"xwin-lm/xwin-lm-70b",
+}
--- a/relay/adaptor/vertexai/gemini/adapter.go
+++ b/relay/adaptor/vertexai/gemini/adapter.go
@@ -16,10 +16,12 @@ import (

 var ModelList = []string{
 	"gemini-pro", "gemini-pro-vision",
-	"gemini-1.5-pro-001", "gemini-1.5-flash-001",
-	"gemini-1.5-pro-002", "gemini-1.5-flash-002",
-	"gemini-2.0-flash-exp",
-	"gemini-2.0-flash-thinking-exp", "gemini-2.0-flash-thinking-exp-01-21",
+	"gemini-exp-1206",
+	"gemini-1.5-pro-001", "gemini-1.5-pro-002",
+	"gemini-1.5-flash-001", "gemini-1.5-flash-002",
+	"gemini-2.0-flash-exp", "gemini-2.0-flash-001",
+	"gemini-2.0-flash-lite-preview-02-05",
+	"gemini-2.0-flash-thinking-exp-01-21",
 }

 type Adaptor struct {
--- a/relay/adaptor/xai/constants.go
+++ b/relay/adaptor/xai/constants.go
@@ -1,5 +1,14 @@
 package xai

+//https://console.x.ai/
+
 var ModelList = []string{
+	"grok-2",
+	"grok-vision-beta",
+	"grok-2-vision-1212",
+	"grok-2-vision",
+	"grok-2-vision-latest",
+	"grok-2-1212",
+	"grok-2-latest",
 	"grok-beta",
 }
--- a/relay/adaptor/xunfei/constants.go
+++ b/relay/adaptor/xunfei/constants.go
@@ -1,12 +1,10 @@
 package xunfei

 var ModelList = []string{
-	"SparkDesk",
-	"SparkDesk-v1.1",
-	"SparkDesk-v2.1",
-	"SparkDesk-v3.1",
-	"SparkDesk-v3.1-128K",
-	"SparkDesk-v3.5",
-	"SparkDesk-v3.5-32K",
-	"SparkDesk-v4.0",
+	"Spark-Lite",
+	"Spark-Pro",
+	"Spark-Pro-128K",
+	"Spark-Max",
+	"Spark-Max-32K",
+	"Spark-4.0-Ultra",
 }
--- a/relay/adaptor/xunfei/domain.go
+++ b/relay/adaptor/xunfei/domain.go
@@ -0,0 +1,97 @@
+package xunfei
+
+import (
+	"fmt"
+	"strings"
+)
+
+// https://www.xfyun.cn/doc/spark/Web.html#_1-%E6%8E%A5%E5%8F%A3%E8%AF%B4%E6%98%8E
+
+//Spark4.0 Ultra 请求地址，对应的domain参数为4.0Ultra：
+//
+//wss://spark-api.xf-yun.com/v4.0/chat
+//Spark Max-32K请求地址，对应的domain参数为max-32k
+//
+//wss://spark-api.xf-yun.com/chat/max-32k
+//Spark Max请求地址，对应的domain参数为generalv3.5
+//
+//wss://spark-api.xf-yun.com/v3.5/chat
+//Spark Pro-128K请求地址，对应的domain参数为pro-128k：
+//
+// wss://spark-api.xf-yun.com/chat/pro-128k
+//Spark Pro请求地址，对应的domain参数为generalv3：
+//
+//wss://spark-api.xf-yun.com/v3.1/chat
+//Spark Lite请求地址，对应的domain参数为lite：
+//
+//wss://spark-api.xf-yun.com/v1.1/chat
+
+// Lite、Pro、Pro-128K、Max、Max-32K和4.0 Ultra
+
+func parseAPIVersionByModelName(modelName string) string {
+	apiVersion := modelName2APIVersion(modelName)
+	if apiVersion != "" {
+		return apiVersion
+	}
+
+	index := strings.IndexAny(modelName, "-")
+	if index != -1 {
+		return modelName[index+1:]
+	}
+	return ""
+}
+
+func modelName2APIVersion(modelName string) string {
+	switch modelName {
+	case "Spark-Lite":
+		return "v1.1"
+	case "Spark-Pro":
+		return "v3.1"
+	case "Spark-Pro-128K":
+		return "v3.1-128K"
+	case "Spark-Max":
+		return "v3.5"
+	case "Spark-Max-32K":
+		return "v3.5-32K"
+	case "Spark-4.0-Ultra":
+		return "v4.0"
+	}
+	return ""
+}
+
+// https://www.xfyun.cn/doc/spark/Web.html#_1-%E6%8E%A5%E5%8F%A3%E8%AF%B4%E6%98%8E
+func apiVersion2domain(apiVersion string) string {
+	switch apiVersion {
+	case "v1.1":
+		return "lite"
+	case "v2.1":
+		return "generalv2"
+	case "v3.1":
+		return "generalv3"
+	case "v3.1-128K":
+		return "pro-128k"
+	case "v3.5":
+		return "generalv3.5"
+	case "v3.5-32K":
+		return "max-32k"
+	case "v4.0":
+		return "4.0Ultra"
+	}
+	return "general" + apiVersion
+}
+
+func getXunfeiAuthUrl(apiVersion string, apiKey string, apiSecret string) (string, string) {
+	var authUrl string
+	domain := apiVersion2domain(apiVersion)
+	switch apiVersion {
+	case "v3.1-128K":
+		authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/chat/pro-128k"), apiKey, apiSecret)
+		break
+	case "v3.5-32K":
+		authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/chat/max-32k"), apiKey, apiSecret)
+		break
+	default:
+		authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/%s/chat", apiVersion), apiKey, apiSecret)
+	}
+	return domain, authUrl
+}
--- a/relay/adaptor/xunfei/main.go
+++ b/relay/adaptor/xunfei/main.go
@@ -15,6 +15,7 @@ import (

 	"github.com/gin-gonic/gin"
 	"github.com/gorilla/websocket"
+
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/logger"
@@ -270,48 +271,3 @@ func xunfeiMakeRequest(textRequest model.GeneralOpenAIRequest, domain, authUrl,

 	return dataChan, stopChan, nil
 }
-
-func parseAPIVersionByModelName(modelName string) string {
-	index := strings.IndexAny(modelName, "-")
-	if index != -1 {
-		return modelName[index+1:]
-	}
-	return ""
-}
-
-// https://www.xfyun.cn/doc/spark/Web.html#_1-%E6%8E%A5%E5%8F%A3%E8%AF%B4%E6%98%8E
-func apiVersion2domain(apiVersion string) string {
-	switch apiVersion {
-	case "v1.1":
-		return "lite"
-	case "v2.1":
-		return "generalv2"
-	case "v3.1":
-		return "generalv3"
-	case "v3.1-128K":
-		return "pro-128k"
-	case "v3.5":
-		return "generalv3.5"
-	case "v3.5-32K":
-		return "max-32k"
-	case "v4.0":
-		return "4.0Ultra"
-	}
-	return "general" + apiVersion
-}
-
-func getXunfeiAuthUrl(apiVersion string, apiKey string, apiSecret string) (string, string) {
-	var authUrl string
-	domain := apiVersion2domain(apiVersion)
-	switch apiVersion {
-	case "v3.1-128K":
-		authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/chat/pro-128k"), apiKey, apiSecret)
-		break
-	case "v3.5-32K":
-		authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/chat/max-32k"), apiKey, apiSecret)
-		break
-	default:
-		authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/%s/chat", apiVersion), apiKey, apiSecret)
-	}
-	return domain, authUrl
-}
--- a/relay/adaptor/xunfeiv2/constants.go
+++ b/relay/adaptor/xunfeiv2/constants.go
@@ -0,0 +1,12 @@
+package xunfeiv2
+
+// https://www.xfyun.cn/doc/spark/HTTP%E8%B0%83%E7%94%A8%E6%96%87%E6%A1%A3.html#_3-%E8%AF%B7%E6%B1%82%E8%AF%B4%E6%98%8E
+
+var ModelList = []string{
+	"lite",
+	"generalv3",
+	"pro-128k",
+	"generalv3.5",
+	"max-32k",
+	"4.0Ultra",
+}
--- a/relay/apitype/define.go
+++ b/relay/apitype/define.go
@@ -20,6 +20,6 @@ const (
 	VertexAI
 	Proxy
 	Replicate
-
+	CozeV3
 	Dummy // this one is only for count, do not add any channel after this
 )
--- a/relay/billing/ratio/group.go
+++ b/relay/billing/ratio/group.go
@@ -3,8 +3,10 @@ package ratio
 import (
 	"encoding/json"
 	"github.com/songquanpeng/one-api/common/logger"
+	"sync"
 )

+var groupRatioLock sync.RWMutex
 var GroupRatio = map[string]float64{
 	"default": 1,
 	"vip":     1,
@@ -20,11 +22,15 @@ func GroupRatio2JSONString() string {
 }

 func UpdateGroupRatioByJSONString(jsonStr string) error {
+	groupRatioLock.Lock()
+	defer groupRatioLock.Unlock()
 	GroupRatio = make(map[string]float64)
 	return json.Unmarshal([]byte(jsonStr), &GroupRatio)
 }

 func GetGroupRatio(name string) float64 {
+	groupRatioLock.RLock()
+	defer groupRatioLock.RUnlock()
 	ratio, ok := GroupRatio[name]
 	if !ok {
 		logger.SysError("group ratio not found: " + name)
--- a/relay/billing/ratio/model.go
+++ b/relay/billing/ratio/model.go
@@ -4,6 +4,7 @@ import (
 	"encoding/json"
 	"fmt"
 	"strings"
+	"sync"

 	"github.com/songquanpeng/one-api/common/logger"
 )
@@ -15,6 +16,8 @@ const (
 	RMB       = USD / USD2RMB
 )

+var modelRatioLock sync.RWMutex
+
 // ModelRatio
 // https://platform.openai.com/docs/models/model-endpoint-compatibility
 // https://cloud.baidu.com/doc/WENXINWORKSHOP/s/Blfmc9dlf
@@ -24,6 +27,7 @@ const (
 var ModelRatio = map[string]float64{
 	// https://openai.com/pricing
 	"gpt-4":                   15,
+	"gpt-4.1":                 15,
 	"gpt-4-0314":              15,
 	"gpt-4-0613":              15,
 	"gpt-4-32k":               30,
@@ -56,6 +60,8 @@ var ModelRatio = map[string]float64{
 	"o1-preview-2024-09-12":   7.5,
 	"o1-mini":                 1.5, // $3.00 / 1M input tokens
 	"o1-mini-2024-09-12":      1.5,
+	"o3-mini":                 1.5, // $3.00 / 1M input tokens
+	"o3-mini-2025-01-31":      1.5,
 	"davinci-002":             1,   // $0.002 / 1K tokens
 	"babbage-002":             0.2, // $0.0004 / 1K tokens
 	"text-ada-001":            0.2,
@@ -66,6 +72,8 @@ var ModelRatio = map[string]float64{
 	"text-davinci-edit-001":   10,
 	"code-davinci-edit-001":   10,
 	"whisper-1":               15,  // $0.006 / minute -> $0.006 / 150 words -> $0.006 / 200 tokens -> $0.03 / 1k tokens
+	"gpt-4o-mini-transcribe":  15,  // $0.006 / minute -> $0.006 / 150 words -> $0.006 / 200 tokens -> $0.03 / 1k tokens
+	"gpt-4o-transcribe":       15,  // $0.006 / minute -> $0.006 / 150 words -> $0.006 / 200 tokens -> $0.03 / 1k tokens
 	"tts-1":                   7.5, // $0.015 / 1K characters
 	"tts-1-1106":              7.5,
 	"tts-1-hd":                15, // $0.030 / 1K characters
@@ -88,11 +96,11 @@ var ModelRatio = map[string]float64{
 	"claude-2.1":                 8.0 / 1000 * USD,
 	"claude-3-haiku-20240307":    0.25 / 1000 * USD,
 	"claude-3-5-haiku-20241022":  1.0 / 1000 * USD,
-	"claude-3-5-haiku-latest":    1.0 / 1000 * USD,	
+	"claude-3-5-haiku-latest":    1.0 / 1000 * USD,
 	"claude-3-sonnet-20240229":   3.0 / 1000 * USD,
 	"claude-3-5-sonnet-20240620": 3.0 / 1000 * USD,
 	"claude-3-5-sonnet-20241022": 3.0 / 1000 * USD,
-	"claude-3-5-sonnet-latest"  : 3.0 / 1000 * USD,	
+	"claude-3-5-sonnet-latest":   3.0 / 1000 * USD,
 	"claude-3-opus-20240229":     15.0 / 1000 * USD,
 	// https://cloud.baidu.com/doc/WENXINWORKSHOP/s/hlrk4akp7
 	"ERNIE-4.0-8K":       0.120 * RMB,
@@ -112,15 +120,24 @@ var ModelRatio = map[string]float64{
 	"bge-large-en":       0.002 * RMB,
 	"tao-8k":             0.002 * RMB,
 	// https://ai.google.dev/pricing
-	"gemini-pro":                          1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
-	"gemini-1.0-pro":                      1,
-	"gemini-1.5-pro":                      1,
-	"gemini-1.5-pro-001":                  1,
-	"gemini-1.5-flash":                    1,
-	"gemini-1.5-flash-001":                1,
-	"gemini-2.0-flash-exp":                1,
-	"gemini-2.0-flash-thinking-exp":       1,
-	"gemini-2.0-flash-thinking-exp-01-21": 1,
+	// https://cloud.google.com/vertex-ai/generative-ai/pricing
+	// "gemma-2-2b-it":                       0,
+	// "gemma-2-9b-it":                       0,
+	// "gemma-2-27b-it":                      0,
+	"gemini-pro":                          0.25 * MILLI_USD, // $0.00025 / 1k characters -> $0.001 / 1k tokens
+	"gemini-1.0-pro":                      0.125 * MILLI_USD,
+	"gemini-1.5-pro":                      1.25 * MILLI_USD,
+	"gemini-1.5-pro-001":                  1.25 * MILLI_USD,
+	"gemini-1.5-pro-experimental":         1.25 * MILLI_USD,
+	"gemini-1.5-flash":                    0.075 * MILLI_USD,
+	"gemini-1.5-flash-001":                0.075 * MILLI_USD,
+	"gemini-1.5-flash-8b":                 0.0375 * MILLI_USD,
+	"gemini-2.0-flash-exp":                0.075 * MILLI_USD,
+	"gemini-2.0-flash":                    0.15 * MILLI_USD,
+	"gemini-2.0-flash-001":                0.15 * MILLI_USD,
+	"gemini-2.0-flash-lite-preview-02-05": 0.075 * MILLI_USD,
+	"gemini-2.0-flash-thinking-exp-01-21": 0.075 * MILLI_USD,
+	"gemini-2.0-pro-exp-02-05":            1.25 * MILLI_USD,
 	"aqa":                                 1,
 	// https://open.bigmodel.cn/pricing
 	"glm-zero-preview": 0.01 * RMB,
@@ -147,91 +164,105 @@ var ModelRatio = map[string]float64{
 	"embedding-2":      0.0005 * RMB,
 	"embedding-3":      0.0005 * RMB,
 	// https://help.aliyun.com/zh/dashscope/developer-reference/tongyi-thousand-questions-metering-and-billing
-	"qwen-turbo":                  1.4286, // ￥0.02 / 1k tokens
-	"qwen-turbo-latest":           1.4286,
-	"qwen-plus":                   1.4286,
-	"qwen-plus-latest":            1.4286,
-	"qwen-max":                    1.4286,
-	"qwen-max-latest":             1.4286,
-	"qwen-max-longcontext":        1.4286,
-	"qwen-vl-max":                 1.4286,
-	"qwen-vl-max-latest":          1.4286,
-	"qwen-vl-plus":                1.4286,
-	"qwen-vl-plus-latest":         1.4286,
-	"qwen-vl-ocr":                 1.4286,
-	"qwen-vl-ocr-latest":          1.4286,
-	"qwen-audio-turbo":            1.4286,
-	"qwen-math-plus":              1.4286,
-	"qwen-math-plus-latest":       1.4286,
-	"qwen-math-turbo":             1.4286,
-	"qwen-math-turbo-latest":      1.4286,
-	"qwen-coder-plus":             1.4286,
-	"qwen-coder-plus-latest":      1.4286,
-	"qwen-coder-turbo":            1.4286,
-	"qwen-coder-turbo-latest":     1.4286,
-	"qwq-32b-preview":             1.4286,
-	"qwen2.5-72b-instruct":        1.4286,
-	"qwen2.5-32b-instruct":        1.4286,
-	"qwen2.5-14b-instruct":        1.4286,
-	"qwen2.5-7b-instruct":         1.4286,
-	"qwen2.5-3b-instruct":         1.4286,
-	"qwen2.5-1.5b-instruct":       1.4286,
-	"qwen2.5-0.5b-instruct":       1.4286,
-	"qwen2-72b-instruct":          1.4286,
-	"qwen2-57b-a14b-instruct":     1.4286,
-	"qwen2-7b-instruct":           1.4286,
-	"qwen2-1.5b-instruct":         1.4286,
-	"qwen2-0.5b-instruct":         1.4286,
-	"qwen1.5-110b-chat":           1.4286,
-	"qwen1.5-72b-chat":            1.4286,
-	"qwen1.5-32b-chat":            1.4286,
-	"qwen1.5-14b-chat":            1.4286,
-	"qwen1.5-7b-chat":             1.4286,
-	"qwen1.5-1.8b-chat":           1.4286,
-	"qwen1.5-0.5b-chat":           1.4286,
-	"qwen-72b-chat":               1.4286,
-	"qwen-14b-chat":               1.4286,
-	"qwen-7b-chat":                1.4286,
-	"qwen-1.8b-chat":              1.4286,
-	"qwen-1.8b-longcontext-chat":  1.4286,
-	"qwen2-vl-7b-instruct":        1.4286,
-	"qwen2-vl-2b-instruct":        1.4286,
-	"qwen-vl-v1":                  1.4286,
-	"qwen-vl-chat-v1":             1.4286,
-	"qwen2-audio-instruct":        1.4286,
-	"qwen-audio-chat":             1.4286,
-	"qwen2.5-math-72b-instruct":   1.4286,
-	"qwen2.5-math-7b-instruct":    1.4286,
-	"qwen2.5-math-1.5b-instruct":  1.4286,
-	"qwen2-math-72b-instruct":     1.4286,
-	"qwen2-math-7b-instruct":      1.4286,
-	"qwen2-math-1.5b-instruct":    1.4286,
-	"qwen2.5-coder-32b-instruct":  1.4286,
-	"qwen2.5-coder-14b-instruct":  1.4286,
-	"qwen2.5-coder-7b-instruct":   1.4286,
-	"qwen2.5-coder-3b-instruct":   1.4286,
-	"qwen2.5-coder-1.5b-instruct": 1.4286,
-	"qwen2.5-coder-0.5b-instruct": 1.4286,
-	"text-embedding-v1":           0.05, // ￥0.0007 / 1k tokens
-	"text-embedding-v3":           0.05,
-	"text-embedding-v2":           0.05,
-	"text-embedding-async-v2":     0.05,
-	"text-embedding-async-v1":     0.05,
-	"ali-stable-diffusion-xl":     8.00,
-	"ali-stable-diffusion-v1.5":   8.00,
-	"wanx-v1":                     8.00,
-	"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
-	"SparkDesk-v3.1":              1.2858, // ￥0.018 / 1k tokens
-	"SparkDesk-v3.1-128K":         1.2858, // ￥0.018 / 1k tokens
-	"SparkDesk-v3.5":              1.2858, // ￥0.018 / 1k tokens
-	"SparkDesk-v3.5-32K":          1.2858, // ￥0.018 / 1k tokens
-	"SparkDesk-v4.0":              1.2858, // ￥0.018 / 1k tokens
-	"360GPT_S2_V9":                0.8572, // ¥0.012 / 1k tokens
-	"embedding-bert-512-v1":       0.0715, // ¥0.001 / 1k tokens
-	"embedding_s1_v1":             0.0715, // ¥0.001 / 1k tokens
-	"semantic_similarity_s1_v1":   0.0715, // ¥0.001 / 1k tokens
+	"qwen-turbo":                    0.0003 * RMB,
+	"qwen-turbo-latest":             0.0003 * RMB,
+	"qwen-plus":                     0.0008 * RMB,
+	"qwen-plus-latest":              0.0008 * RMB,
+	"qwen-max":                      0.0024 * RMB,
+	"qwen-max-latest":               0.0024 * RMB,
+	"qwen-max-longcontext":          0.0005 * RMB,
+	"qwen-vl-max":                   0.003 * RMB,
+	"qwen-vl-max-latest":            0.003 * RMB,
+	"qwen-vl-plus":                  0.0015 * RMB,
+	"qwen-vl-plus-latest":           0.0015 * RMB,
+	"qwen-vl-ocr":                   0.005 * RMB,
+	"qwen-vl-ocr-latest":            0.005 * RMB,
+	"qwen-audio-turbo":              1.4286,
+	"qwen-math-plus":                0.004 * RMB,
+	"qwen-math-plus-latest":         0.004 * RMB,
+	"qwen-math-turbo":               0.002 * RMB,
+	"qwen-math-turbo-latest":        0.002 * RMB,
+	"qwen-coder-plus":               0.0035 * RMB,
+	"qwen-coder-plus-latest":        0.0035 * RMB,
+	"qwen-coder-turbo":              0.002 * RMB,
+	"qwen-coder-turbo-latest":       0.002 * RMB,
+	"qwen-mt-plus":                  0.015 * RMB,
+	"qwen-mt-turbo":                 0.001 * RMB,
+	"qwq-32b-preview":               0.002 * RMB,
+	"qwen2.5-72b-instruct":          0.004 * RMB,
+	"qwen2.5-32b-instruct":          0.03 * RMB,
+	"qwen2.5-14b-instruct":          0.001 * RMB,
+	"qwen2.5-7b-instruct":           0.0005 * RMB,
+	"qwen2.5-3b-instruct":           0.006 * RMB,
+	"qwen2.5-1.5b-instruct":         0.0003 * RMB,
+	"qwen2.5-0.5b-instruct":         0.0003 * RMB,
+	"qwen2-72b-instruct":            0.004 * RMB,
+	"qwen2-57b-a14b-instruct":       0.0035 * RMB,
+	"qwen2-7b-instruct":             0.001 * RMB,
+	"qwen2-1.5b-instruct":           0.001 * RMB,
+	"qwen2-0.5b-instruct":           0.001 * RMB,
+	"qwen1.5-110b-chat":             0.007 * RMB,
+	"qwen1.5-72b-chat":              0.005 * RMB,
+	"qwen1.5-32b-chat":              0.0035 * RMB,
+	"qwen1.5-14b-chat":              0.002 * RMB,
+	"qwen1.5-7b-chat":               0.001 * RMB,
+	"qwen1.5-1.8b-chat":             0.001 * RMB,
+	"qwen1.5-0.5b-chat":             0.001 * RMB,
+	"qwen-72b-chat":                 0.02 * RMB,
+	"qwen-14b-chat":                 0.008 * RMB,
+	"qwen-7b-chat":                  0.006 * RMB,
+	"qwen-1.8b-chat":                0.006 * RMB,
+	"qwen-1.8b-longcontext-chat":    0.006 * RMB,
+	"qvq-72b-preview":               0.012 * RMB,
+	"qwen2.5-vl-72b-instruct":       0.016 * RMB,
+	"qwen2.5-vl-7b-instruct":        0.002 * RMB,
+	"qwen2.5-vl-3b-instruct":        0.0012 * RMB,
+	"qwen2-vl-7b-instruct":          0.016 * RMB,
+	"qwen2-vl-2b-instruct":          0.002 * RMB,
+	"qwen-vl-v1":                    0.002 * RMB,
+	"qwen-vl-chat-v1":               0.002 * RMB,
+	"qwen2-audio-instruct":          0.002 * RMB,
+	"qwen-audio-chat":               0.002 * RMB,
+	"qwen2.5-math-72b-instruct":     0.004 * RMB,
+	"qwen2.5-math-7b-instruct":      0.001 * RMB,
+	"qwen2.5-math-1.5b-instruct":    0.001 * RMB,
+	"qwen2-math-72b-instruct":       0.004 * RMB,
+	"qwen2-math-7b-instruct":        0.001 * RMB,
+	"qwen2-math-1.5b-instruct":      0.001 * RMB,
+	"qwen2.5-coder-32b-instruct":    0.002 * RMB,
+	"qwen2.5-coder-14b-instruct":    0.002 * RMB,
+	"qwen2.5-coder-7b-instruct":     0.001 * RMB,
+	"qwen2.5-coder-3b-instruct":     0.001 * RMB,
+	"qwen2.5-coder-1.5b-instruct":   0.001 * RMB,
+	"qwen2.5-coder-0.5b-instruct":   0.001 * RMB,
+	"text-embedding-v1":             0.0007 * RMB, // ￥0.0007 / 1k tokens
+	"text-embedding-v3":             0.0007 * RMB,
+	"text-embedding-v2":             0.0007 * RMB,
+	"text-embedding-async-v2":       0.0007 * RMB,
+	"text-embedding-async-v1":       0.0007 * RMB,
+	"ali-stable-diffusion-xl":       8.00,
+	"ali-stable-diffusion-v1.5":     8.00,
+	"wanx-v1":                       8.00,
+	"deepseek-r1":                   0.002 * RMB,
+	"deepseek-v3":                   0.001 * RMB,
+	"deepseek-r1-distill-qwen-1.5b": 0.001 * RMB,
+	"deepseek-r1-distill-qwen-7b":   0.0005 * RMB,
+	"deepseek-r1-distill-qwen-14b":  0.001 * RMB,
+	"deepseek-r1-distill-qwen-32b":  0.002 * RMB,
+	"deepseek-r1-distill-llama-8b":  0.0005 * RMB,
+	"deepseek-r1-distill-llama-70b": 0.004 * RMB,
+	"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
+	"SparkDesk-v3.1":                1.2858, // ￥0.018 / 1k tokens
+	"SparkDesk-v3.1-128K":           1.2858, // ￥0.018 / 1k tokens
+	"SparkDesk-v3.5":                1.2858, // ￥0.018 / 1k tokens
+	"SparkDesk-v3.5-32K":            1.2858, // ￥0.018 / 1k tokens
+	"SparkDesk-v4.0":                1.2858, // ￥0.018 / 1k tokens
+	"360GPT_S2_V9":                  0.8572, // ¥0.012 / 1k tokens
+	"embedding-bert-512-v1":         0.0715, // ¥0.001 / 1k tokens
+	"embedding_s1_v1":               0.0715, // ¥0.001 / 1k tokens
+	"semantic_similarity_s1_v1":     0.0715, // ¥0.001 / 1k tokens
 	// https://cloud.tencent.com/document/product/1729/97731#e0e6be58-60c8-469f-bdeb-6c264ce3b4d0
 	"hunyuan-turbo":             0.015 * RMB,
 	"hunyuan-large":             0.004 * RMB,
@@ -359,6 +390,238 @@ var ModelRatio = map[string]float64{
 	"mistralai/mistral-7b-instruct-v0.2":        0.050 * USD,
 	"mistralai/mistral-7b-v0.1":                 0.050 * USD,
 	"mistralai/mixtral-8x7b-instruct-v0.1":      0.300 * USD,
+	//https://openrouter.ai/models
+	"01-ai/yi-large":                                  1.5,
+	"aetherwiing/mn-starcannon-12b":                   0.6,
+	"ai21/jamba-1-5-large":                            4.0,
+	"ai21/jamba-1-5-mini":                             0.2,
+	"ai21/jamba-instruct":                             0.35,
+	"aion-labs/aion-1.0":                              6.0,
+	"aion-labs/aion-1.0-mini":                         1.2,
+	"aion-labs/aion-rp-llama-3.1-8b":                  0.1,
+	"allenai/llama-3.1-tulu-3-405b":                   5.0,
+	"alpindale/goliath-120b":                          4.6875,
+	"alpindale/magnum-72b":                            1.125,
+	"amazon/nova-lite-v1":                             0.12,
+	"amazon/nova-micro-v1":                            0.07,
+	"amazon/nova-pro-v1":                              1.6,
+	"anthracite-org/magnum-v2-72b":                    1.5,
+	"anthracite-org/magnum-v4-72b":                    1.125,
+	"anthropic/claude-2":                              12.0,
+	"anthropic/claude-2.0":                            12.0,
+	"anthropic/claude-2.0:beta":                       12.0,
+	"anthropic/claude-2.1":                            12.0,
+	"anthropic/claude-2.1:beta":                       12.0,
+	"anthropic/claude-2:beta":                         12.0,
+	"anthropic/claude-3-haiku":                        0.625,
+	"anthropic/claude-3-haiku:beta":                   0.625,
+	"anthropic/claude-3-opus":                         37.5,
+	"anthropic/claude-3-opus:beta":                    37.5,
+	"anthropic/claude-3-sonnet":                       7.5,
+	"anthropic/claude-3-sonnet:beta":                  7.5,
+	"anthropic/claude-3.5-haiku":                      2.0,
+	"anthropic/claude-3.5-haiku-20241022":             2.0,
+	"anthropic/claude-3.5-haiku-20241022:beta":        2.0,
+	"anthropic/claude-3.5-haiku:beta":                 2.0,
+	"anthropic/claude-3.5-sonnet":                     7.5,
+	"anthropic/claude-3.5-sonnet-20240620":            7.5,
+	"anthropic/claude-3.5-sonnet-20240620:beta":       7.5,
+	"anthropic/claude-3.5-sonnet:beta":                7.5,
+	"cognitivecomputations/dolphin-mixtral-8x22b":     0.45,
+	"cognitivecomputations/dolphin-mixtral-8x7b":      0.25,
+	"cohere/command":                                  0.95,
+	"cohere/command-r":                                0.7125,
+	"cohere/command-r-03-2024":                        0.7125,
+	"cohere/command-r-08-2024":                        0.285,
+	"cohere/command-r-plus":                           7.125,
+	"cohere/command-r-plus-04-2024":                   7.125,
+	"cohere/command-r-plus-08-2024":                   4.75,
+	"cohere/command-r7b-12-2024":                      0.075,
+	"databricks/dbrx-instruct":                        0.6,
+	"deepseek/deepseek-chat":                          0.445,
+	"deepseek/deepseek-chat-v2.5":                     1.0,
+	"deepseek/deepseek-chat:free":                     0.0,
+	"deepseek/deepseek-r1":                            1.2,
+	"deepseek/deepseek-r1-distill-llama-70b":          0.345,
+	"deepseek/deepseek-r1-distill-llama-70b:free":     0.0,
+	"deepseek/deepseek-r1-distill-llama-8b":           0.02,
+	"deepseek/deepseek-r1-distill-qwen-1.5b":          0.09,
+	"deepseek/deepseek-r1-distill-qwen-14b":           0.075,
+	"deepseek/deepseek-r1-distill-qwen-32b":           0.09,
+	"deepseek/deepseek-r1:free":                       0.0,
+	"eva-unit-01/eva-llama-3.33-70b":                  3.0,
+	"eva-unit-01/eva-qwen-2.5-32b":                    1.7,
+	"eva-unit-01/eva-qwen-2.5-72b":                    3.0,
+	"google/gemini-2.0-flash-001":                     0.2,
+	"google/gemini-2.0-flash-exp:free":                0.0,
+	"google/gemini-2.0-flash-lite-preview-02-05:free": 0.0,
+	"google/gemini-2.0-flash-thinking-exp-1219:free":  0.0,
+	"google/gemini-2.0-flash-thinking-exp:free":       0.0,
+	"google/gemini-2.0-pro-exp-02-05:free":            0.0,
+	"google/gemini-exp-1206:free":                     0.0,
+	"google/gemini-flash-1.5":                         0.15,
+	"google/gemini-flash-1.5-8b":                      0.075,
+	"google/gemini-flash-1.5-8b-exp":                  0.0,
+	"google/gemini-pro":                               0.75,
+	"google/gemini-pro-1.5":                           2.5,
+	"google/gemini-pro-vision":                        0.75,
+	"google/gemma-2-27b-it":                           0.135,
+	"google/gemma-2-9b-it":                            0.03,
+	"google/gemma-2-9b-it:free":                       0.0,
+	"google/gemma-7b-it":                              0.075,
+	"google/learnlm-1.5-pro-experimental:free":        0.0,
+	"google/palm-2-chat-bison":                        1.0,
+	"google/palm-2-chat-bison-32k":                    1.0,
+	"google/palm-2-codechat-bison":                    1.0,
+	"google/palm-2-codechat-bison-32k":                1.0,
+	"gryphe/mythomax-l2-13b":                          0.0325,
+	"gryphe/mythomax-l2-13b:free":                     0.0,
+	"huggingfaceh4/zephyr-7b-beta:free":               0.0,
+	"infermatic/mn-inferor-12b":                       0.6,
+	"inflection/inflection-3-pi":                      5.0,
+	"inflection/inflection-3-productivity":            5.0,
+	"jondurbin/airoboros-l2-70b":                      0.25,
+	"liquid/lfm-3b":                                   0.01,
+	"liquid/lfm-40b":                                  0.075,
+	"liquid/lfm-7b":                                   0.005,
+	"mancer/weaver":                                   1.125,
+	"meta-llama/llama-2-13b-chat":                     0.11,
+	"meta-llama/llama-2-70b-chat":                     0.45,
+	"meta-llama/llama-3-70b-instruct":                 0.2,
+	"meta-llama/llama-3-8b-instruct":                  0.03,
+	"meta-llama/llama-3-8b-instruct:free":             0.0,
+	"meta-llama/llama-3.1-405b":                       1.0,
+	"meta-llama/llama-3.1-405b-instruct":              0.4,
+	"meta-llama/llama-3.1-70b-instruct":               0.15,
+	"meta-llama/llama-3.1-8b-instruct":                0.025,
+	"meta-llama/llama-3.2-11b-vision-instruct":        0.0275,
+	"meta-llama/llama-3.2-11b-vision-instruct:free":   0.0,
+	"meta-llama/llama-3.2-1b-instruct":                0.005,
+	"meta-llama/llama-3.2-3b-instruct":                0.0125,
+	"meta-llama/llama-3.2-90b-vision-instruct":        0.8,
+	"meta-llama/llama-3.3-70b-instruct":               0.15,
+	"meta-llama/llama-3.3-70b-instruct:free":          0.0,
+	"meta-llama/llama-guard-2-8b":                     0.1,
+	"microsoft/phi-3-medium-128k-instruct":            0.5,
+	"microsoft/phi-3-medium-128k-instruct:free":       0.0,
+	"microsoft/phi-3-mini-128k-instruct":              0.05,
+	"microsoft/phi-3-mini-128k-instruct:free":         0.0,
+	"microsoft/phi-3.5-mini-128k-instruct":            0.05,
+	"microsoft/phi-4":                                 0.07,
+	"microsoft/wizardlm-2-7b":                         0.035,
+	"microsoft/wizardlm-2-8x22b":                      0.25,
+	"minimax/minimax-01":                              0.55,
+	"mistralai/codestral-2501":                        0.45,
+	"mistralai/codestral-mamba":                       0.125,
+	"mistralai/ministral-3b":                          0.02,
+	"mistralai/ministral-8b":                          0.05,
+	"mistralai/mistral-7b-instruct":                   0.0275,
+	"mistralai/mistral-7b-instruct-v0.1":              0.1,
+	"mistralai/mistral-7b-instruct-v0.3":              0.0275,
+	"mistralai/mistral-7b-instruct:free":              0.0,
+	"mistralai/mistral-large":                         3.0,
+	"mistralai/mistral-large-2407":                    3.0,
+	"mistralai/mistral-large-2411":                    3.0,
+	"mistralai/mistral-medium":                        4.05,
+	"mistralai/mistral-nemo":                          0.04,
+	"mistralai/mistral-nemo:free":                     0.0,
+	"mistralai/mistral-small":                         0.3,
+	"mistralai/mistral-small-24b-instruct-2501":       0.07,
+	"mistralai/mistral-small-24b-instruct-2501:free":  0.0,
+	"mistralai/mistral-tiny":                          0.125,
+	"mistralai/mixtral-8x22b-instruct":                0.45,
+	"mistralai/mixtral-8x7b":                          0.3,
+	"mistralai/mixtral-8x7b-instruct":                 0.12,
+	"mistralai/pixtral-12b":                           0.05,
+	"mistralai/pixtral-large-2411":                    3.0,
+	"neversleep/llama-3-lumimaid-70b":                 2.25,
+	"neversleep/llama-3-lumimaid-8b":                  0.5625,
+	"neversleep/llama-3-lumimaid-8b:extended":         0.5625,
+	"neversleep/llama-3.1-lumimaid-70b":               2.25,
+	"neversleep/llama-3.1-lumimaid-8b":                0.5625,
+	"neversleep/noromaid-20b":                         1.125,
+	"nothingiisreal/mn-celeste-12b":                   0.6,
+	"nousresearch/hermes-2-pro-llama-3-8b":            0.02,
+	"nousresearch/hermes-3-llama-3.1-405b":            0.4,
+	"nousresearch/hermes-3-llama-3.1-70b":             0.15,
+	"nousresearch/nous-hermes-2-mixtral-8x7b-dpo":     0.3,
+	"nousresearch/nous-hermes-llama2-13b":             0.085,
+	"nvidia/llama-3.1-nemotron-70b-instruct":          0.15,
+	"nvidia/llama-3.1-nemotron-70b-instruct:free":     0.0,
+	"openai/chatgpt-4o-latest":                        7.5,
+	"openai/gpt-3.5-turbo":                            0.75,
+	"openai/gpt-3.5-turbo-0125":                       0.75,
+	"openai/gpt-3.5-turbo-0613":                       1.0,
+	"openai/gpt-3.5-turbo-1106":                       1.0,
+	"openai/gpt-3.5-turbo-16k":                        2.0,
+	"openai/gpt-3.5-turbo-instruct":                   1.0,
+	"openai/gpt-4":                                    30.0,
+	"openai/gpt-4-0314":                               30.0,
+	"openai/gpt-4-1106-preview":                       15.0,
+	"openai/gpt-4-32k":                                60.0,
+	"openai/gpt-4-32k-0314":                           60.0,
+	"openai/gpt-4-turbo":                              15.0,
+	"openai/gpt-4-turbo-preview":                      15.0,
+	"openai/gpt-4o":                                   5.0,
+	"openai/gpt-4o-2024-05-13":                        7.5,
+	"openai/gpt-4o-2024-08-06":                        5.0,
+	"openai/gpt-4o-2024-11-20":                        5.0,
+	"openai/gpt-4o-mini":                              0.3,
+	"openai/gpt-4o-mini-2024-07-18":                   0.3,
+	"openai/gpt-4o:extended":                          9.0,
+	"openai/o1":                                       30.0,
+	"openai/o1-mini":                                  2.2,
+	"openai/o1-mini-2024-09-12":                       2.2,
+	"openai/o1-preview":                               30.0,
+	"openai/o1-preview-2024-09-12":                    30.0,
+	"openai/o3-mini":                                  2.2,
+	"openai/o3-mini-high":                             2.2,
+	"openchat/openchat-7b":                            0.0275,
+	"openchat/openchat-7b:free":                       0.0,
+	"openrouter/auto":                                 -500000.0,
+	"perplexity/llama-3.1-sonar-huge-128k-online":     2.5,
+	"perplexity/llama-3.1-sonar-large-128k-chat":      0.5,
+	"perplexity/llama-3.1-sonar-large-128k-online":    0.5,
+	"perplexity/llama-3.1-sonar-small-128k-chat":      0.1,
+	"perplexity/llama-3.1-sonar-small-128k-online":    0.1,
+	"perplexity/sonar":                                0.5,
+	"perplexity/sonar-reasoning":                      2.5,
+	"pygmalionai/mythalion-13b":                       0.6,
+	"qwen/qvq-72b-preview":                            0.25,
+	"qwen/qwen-2-72b-instruct":                        0.45,
+	"qwen/qwen-2-7b-instruct":                         0.027,
+	"qwen/qwen-2-7b-instruct:free":                    0.0,
+	"qwen/qwen-2-vl-72b-instruct":                     0.2,
+	"qwen/qwen-2-vl-7b-instruct":                      0.05,
+	"qwen/qwen-2.5-72b-instruct":                      0.2,
+	"qwen/qwen-2.5-7b-instruct":                       0.025,
+	"qwen/qwen-2.5-coder-32b-instruct":                0.08,
+	"qwen/qwen-max":                                   3.2,
+	"qwen/qwen-plus":                                  0.6,
+	"qwen/qwen-turbo":                                 0.1,
+	"qwen/qwen-vl-plus:free":                          0.0,
+	"qwen/qwen2.5-vl-72b-instruct:free":               0.0,
+	"qwen/qwq-32b-preview":                            0.09,
+	"raifle/sorcererlm-8x22b":                         2.25,
+	"sao10k/fimbulvetr-11b-v2":                        0.6,
+	"sao10k/l3-euryale-70b":                           0.4,
+	"sao10k/l3-lunaris-8b":                            0.03,
+	"sao10k/l3.1-70b-hanami-x1":                       1.5,
+	"sao10k/l3.1-euryale-70b":                         0.4,
+	"sao10k/l3.3-euryale-70b":                         0.4,
+	"sophosympatheia/midnight-rose-70b":               0.4,
+	"sophosympatheia/rogue-rose-103b-v0.2:free":       0.0,
+	"teknium/openhermes-2.5-mistral-7b":               0.085,
+	"thedrummer/rocinante-12b":                        0.25,
+	"thedrummer/unslopnemo-12b":                       0.25,
+	"undi95/remm-slerp-l2-13b":                        0.6,
+	"undi95/toppy-m-7b":                               0.035,
+	"undi95/toppy-m-7b:free":                          0.0,
+	"x-ai/grok-2-1212":                                5.0,
+	"x-ai/grok-2-vision-1212":                         5.0,
+	"x-ai/grok-beta":                                  7.5,
+	"x-ai/grok-vision-beta":                           7.5,
+	"xwin-lm/xwin-lm-70b":                             1.875,
 }

 var CompletionRatio = map[string]float64{
@@ -366,7 +629,9 @@ var CompletionRatio = map[string]float64{
 	"llama3-8b-8192(33)":  0.0006 / 0.0003,
 	"llama3-70b-8192(33)": 0.0035 / 0.00265,
 	// whisper
-	"whisper-1": 0, // only count input tokens
+	"whisper-1":              0, // only count input tokens
+	"gpt-4o-mini-transcribe": 0,
+	"gpt-4o-transcribe":      0,
 	// deepseek
 	"deepseek-chat":     0.28 / 0.14,
 	"deepseek-reasoner": 2.19 / 0.55,
@@ -417,11 +682,15 @@ func ModelRatio2JSONString() string {
 }

 func UpdateModelRatioByJSONString(jsonStr string) error {
+	modelRatioLock.Lock()
+	defer modelRatioLock.Unlock()
 	ModelRatio = make(map[string]float64)
 	return json.Unmarshal([]byte(jsonStr), &ModelRatio)
 }

 func GetModelRatio(name string, channelType int) float64 {
+	modelRatioLock.RLock()
+	defer modelRatioLock.RUnlock()
 	if strings.HasPrefix(name, "qwen-") && strings.HasSuffix(name, "-internet") {
 		name = strings.TrimSuffix(name, "-internet")
 	}
--- a/relay/channeltype/define.go
+++ b/relay/channeltype/define.go
@@ -48,5 +48,11 @@ const (
 	SiliconFlow
 	XAI
 	Replicate
+	BaiduV2
+	XunfeiV2
+	AliBailian
+	OpenAICompatible
+	GeminiOpenAICompatible
+	CozeV3
 	Dummy
 )
--- a/relay/channeltype/helper.go
+++ b/relay/channeltype/helper.go
@@ -23,12 +23,16 @@ func ToAPIType(channelType int) int {
 		apiType = apitype.Tencent
 	case Gemini:
 		apiType = apitype.Gemini
+	case GeminiOpenAICompatible:
+		apiType = apitype.Gemini
 	case Ollama:
 		apiType = apitype.Ollama
 	case AwsClaude:
 		apiType = apitype.AwsClaude
 	case Coze:
 		apiType = apitype.Coze
+	case CozeV3:
+		apiType = apitype.CozeV3
 	case Cohere:
 		apiType = apitype.Cohere
 	case Cloudflare:
--- a/relay/channeltype/url.go
+++ b/relay/channeltype/url.go
@@ -48,6 +48,13 @@ var ChannelBaseURLs = []string{
 	"https://api.siliconflow.cn",                // 44
 	"https://api.x.ai",                          // 45
 	"https://api.replicate.com/v1/models/",      // 46
+	"https://qianfan.baidubce.com",              // 47
+	"https://spark-api-open.xf-yun.com",         // 48
+	"https://dashscope.aliyuncs.com",            // 49
+	"",                                          // 50
+
+	"https://generativelanguage.googleapis.com/v1beta/openai/", // 51
+	"https://api.coze.cn", // 52
 }

 func init() {
--- a/relay/controller/audio.go
+++ b/relay/controller/audio.go
@@ -8,6 +8,7 @@ import (
 	"errors"
 	"fmt"
 	"io"
+	"mime/multipart"
 	"net/http"
 	"strings"

@@ -30,8 +31,7 @@ import (
 func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatusCode {
 	ctx := c.Request.Context()
 	meta := meta.GetByContext(c)
-	audioModel := "whisper-1"
-
+	audioModel := "gpt-4o-transcribe"
 	tokenId := c.GetInt(ctxkey.TokenId)
 	channelType := c.GetInt(ctxkey.Channel)
 	channelId := c.GetInt(ctxkey.ChannelId)
@@ -124,12 +124,13 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 	fullRequestURL := openai.GetFullRequestURL(baseURL, requestURL, channelType)
 	if channelType == channeltype.Azure {
 		apiVersion := meta.Config.APIVersion
+		deploymentName := c.GetString(ctxkey.ChannelName)
 		if relayMode == relaymode.AudioTranscription {
 			// 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)
+			fullRequestURL = fmt.Sprintf("%s/openai/deployments/%s/audio/transcriptions?api-version=%s", baseURL, deploymentName, apiVersion)
 		} else if relayMode == relaymode.AudioSpeech {
 			// 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)
+			fullRequestURL = fmt.Sprintf("%s/openai/deployments/%s/audio/speech?api-version=%s", baseURL, deploymentName, apiVersion)
 		}
 	}

@@ -138,8 +139,73 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 	if err != nil {
 		return openai.ErrorWrapper(err, "new_request_body_failed", http.StatusInternalServerError)
 	}
-	c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody.Bytes()))
-	responseFormat := c.DefaultPostForm("response_format", "json")
+
+	// 处理表单数据
+	contentType := c.Request.Header.Get("Content-Type")
+	responseFormat := "json"
+	var contentTypeWithBoundary string
+
+	if strings.Contains(contentType, "multipart/form-data") {
+		originalBody := requestBody.Bytes()
+		c.Request.Body = io.NopCloser(bytes.NewBuffer(originalBody))
+		err = c.Request.ParseMultipartForm(32 << 20) // 32MB 最大内存
+		if err != nil {
+			return openai.ErrorWrapper(err, "parse_multipart_form_failed", http.StatusInternalServerError)
+		}
+
+		// 获取响应格式
+		if format := c.Request.FormValue("response_format"); format != "" {
+			responseFormat = format
+		}
+
+		requestBody = &bytes.Buffer{}
+		writer := multipart.NewWriter(requestBody)
+
+		// 复制表单字段
+		for key, values := range c.Request.MultipartForm.Value {
+			for _, value := range values {
+				err = writer.WriteField(key, value)
+				if err != nil {
+					return openai.ErrorWrapper(err, "write_field_failed", http.StatusInternalServerError)
+				}
+			}
+		}
+
+		// 复制文件
+		for key, fileHeaders := range c.Request.MultipartForm.File {
+			for _, fileHeader := range fileHeaders {
+				file, err := fileHeader.Open()
+				if err != nil {
+					return openai.ErrorWrapper(err, "open_file_failed", http.StatusInternalServerError)
+				}
+
+				part, err := writer.CreateFormFile(key, fileHeader.Filename)
+				if err != nil {
+					file.Close()
+					return openai.ErrorWrapper(err, "create_form_file_failed", http.StatusInternalServerError)
+				}
+
+				_, err = io.Copy(part, file)
+				file.Close()
+				if err != nil {
+					return openai.ErrorWrapper(err, "copy_file_failed", http.StatusInternalServerError)
+				}
+			}
+		}
+
+		// 完成multipart写入
+		err = writer.Close()
+		if err != nil {
+			return openai.ErrorWrapper(err, "close_writer_failed", http.StatusInternalServerError)
+		}
+
+		// 更新Content-Type
+		contentTypeWithBoundary = writer.FormDataContentType()
+		c.Request.Header.Set("Content-Type", contentTypeWithBoundary)
+	} else {
+		// 对于非表单请求，直接重置请求体
+		c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody.Bytes()))
+	}

 	req, err := http.NewRequest(c.Request.Method, fullRequestURL, requestBody)
 	if err != nil {
@@ -151,11 +217,26 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		apiKey := c.Request.Header.Get("Authorization")
 		apiKey = strings.TrimPrefix(apiKey, "Bearer ")
 		req.Header.Set("api-key", apiKey)
-		req.ContentLength = c.Request.ContentLength
+		// 确保请求体大小与Content-Length一致
+		req.ContentLength = int64(requestBody.Len())
 	} else {
 		req.Header.Set("Authorization", c.Request.Header.Get("Authorization"))
+		// 确保请求体大小与Content-Length一致
+		req.ContentLength = int64(requestBody.Len())
+	}
+
+	// 确保Content-Type正确传递
+	if strings.Contains(contentType, "multipart/form-data") && c.Request.MultipartForm != nil {
+		// 对于multipart请求，使用我们重建时生成的Content-Type
+		// 注意：此处必须使用writer生成的boundary
+		if contentTypeWithBoundary != "" {
+			req.Header.Set("Content-Type", contentTypeWithBoundary)
+		} else {
+			req.Header.Set("Content-Type", c.Request.Header.Get("Content-Type"))
+		}
+	} else {
+		req.Header.Set("Content-Type", c.Request.Header.Get("Content-Type"))
 	}
-	req.Header.Set("Content-Type", c.Request.Header.Get("Content-Type"))
 	req.Header.Set("Accept", c.Request.Header.Get("Accept"))

 	resp, err := client.HTTPClient.Do(req)
--- a/relay/controller/text.go
+++ b/relay/controller/text.go
@@ -38,7 +38,7 @@ func RelayTextHelper(c *gin.Context) *model.ErrorWithStatusCode {
 	textRequest.Model, _ = getMappedModelName(textRequest.Model, meta.ModelMapping)
 	meta.ActualModelName = textRequest.Model
 	// set system prompt if not empty
-	systemPromptReset := setSystemPrompt(ctx, textRequest, meta.SystemPrompt)
+	systemPromptReset := setSystemPrompt(ctx, textRequest, meta.ForcedSystemPrompt)
 	// get model ratio & group ratio
 	modelRatio := billingratio.GetModelRatio(textRequest.Model, meta.ChannelType)
 	groupRatio := billingratio.GetGroupRatio(meta.Group)
@@ -88,7 +88,11 @@ func RelayTextHelper(c *gin.Context) *model.ErrorWithStatusCode {
 }

 func getRequestBody(c *gin.Context, meta *meta.Meta, textRequest *model.GeneralOpenAIRequest, adaptor adaptor.Adaptor) (io.Reader, error) {
-	if !config.EnforceIncludeUsage && meta.APIType == apitype.OpenAI && meta.OriginModelName == meta.ActualModelName && meta.ChannelType != channeltype.Baichuan {
+	if !config.EnforceIncludeUsage &&
+		meta.APIType == apitype.OpenAI &&
+		meta.OriginModelName == meta.ActualModelName &&
+		meta.ChannelType != channeltype.Baichuan &&
+		meta.ForcedSystemPrompt == "" {
 		// no need to convert request for openai
 		return c.Request.Body, nil
 	}
--- a/relay/meta/relay_meta.go
+++ b/relay/meta/relay_meta.go
@@ -30,29 +30,29 @@ type Meta struct {
 	// OriginModelName is the model name from the raw user request
 	OriginModelName string
 	// ActualModelName is the model name after mapping
-	ActualModelName string
-	RequestURLPath  string
-	PromptTokens    int // only for DoResponse
-	SystemPrompt    string
-	StartTime       time.Time
+	ActualModelName    string
+	RequestURLPath     string
+	PromptTokens       int // only for DoResponse
+	ForcedSystemPrompt string
+	StartTime          time.Time
 }

 func GetByContext(c *gin.Context) *Meta {
 	meta := Meta{
-		Mode:            relaymode.GetByPath(c.Request.URL.Path),
-		ChannelType:     c.GetInt(ctxkey.Channel),
-		ChannelId:       c.GetInt(ctxkey.ChannelId),
-		TokenId:         c.GetInt(ctxkey.TokenId),
-		TokenName:       c.GetString(ctxkey.TokenName),
-		UserId:          c.GetInt(ctxkey.Id),
-		Group:           c.GetString(ctxkey.Group),
-		ModelMapping:    c.GetStringMapString(ctxkey.ModelMapping),
-		OriginModelName: c.GetString(ctxkey.RequestModel),
-		BaseURL:         c.GetString(ctxkey.BaseURL),
-		APIKey:          strings.TrimPrefix(c.Request.Header.Get("Authorization"), "Bearer "),
-		RequestURLPath:  c.Request.URL.String(),
-		SystemPrompt:    c.GetString(ctxkey.SystemPrompt),
-		StartTime:       time.Now(),
+		Mode:               relaymode.GetByPath(c.Request.URL.Path),
+		ChannelType:        c.GetInt(ctxkey.Channel),
+		ChannelId:          c.GetInt(ctxkey.ChannelId),
+		TokenId:            c.GetInt(ctxkey.TokenId),
+		TokenName:          c.GetString(ctxkey.TokenName),
+		UserId:             c.GetInt(ctxkey.Id),
+		Group:              c.GetString(ctxkey.Group),
+		ModelMapping:       c.GetStringMapString(ctxkey.ModelMapping),
+		OriginModelName:    c.GetString(ctxkey.RequestModel),
+		BaseURL:            c.GetString(ctxkey.BaseURL),
+		APIKey:             strings.TrimPrefix(c.Request.Header.Get("Authorization"), "Bearer "),
+		RequestURLPath:     c.Request.URL.String(),
+		ForcedSystemPrompt: c.GetString(ctxkey.SystemPrompt),
+		StartTime:          time.Now(),
 	}
 	cfg, ok := c.Get(ctxkey.Config)
 	if ok {
--- a/relay/model/constant.go
+++ b/relay/model/constant.go
@@ -4,4 +4,5 @@ const (
 	ContentTypeText       = "text"
 	ContentTypeImageURL   = "image_url"
 	ContentTypeInputAudio = "input_audio"
+	ContentTypeInputFile  = "file"
 )
--- a/relay/model/general.go
+++ b/relay/model/general.go
@@ -26,6 +26,7 @@ type GeneralOpenAIRequest struct {
 	Messages            []Message       `json:"messages,omitempty"`
 	Model               string          `json:"model,omitempty"`
 	Store               *bool           `json:"store,omitempty"`
+	ReasoningEffort     *string         `json:"reasoning_effort,omitempty"`
 	Metadata            any             `json:"metadata,omitempty"`
 	FrequencyPenalty    *float64        `json:"frequency_penalty,omitempty"`
 	LogitBias           any             `json:"logit_bias,omitempty"`
--- a/relay/model/message.go
+++ b/relay/model/message.go
@@ -1,11 +1,14 @@
 package model

+import "encoding/json"
+
 type Message struct {
-	Role       string  `json:"role,omitempty"`
-	Content    any     `json:"content,omitempty"`
-	Name       *string `json:"name,omitempty"`
-	ToolCalls  []Tool  `json:"tool_calls,omitempty"`
-	ToolCallId string  `json:"tool_call_id,omitempty"`
+	Role             string  `json:"role,omitempty"`
+	Content          any     `json:"content,omitempty"`
+	ReasoningContent any     `json:"reasoning_content,omitempty"`
+	Name             *string `json:"name,omitempty"`
+	ToolCalls        []Tool  `json:"tool_calls,omitempty"`
+	ToolCallId       string  `json:"tool_call_id,omitempty"`
 }

 func (m Message) IsStringContent() bool {
@@ -37,6 +40,53 @@ func (m Message) StringContent() string {
 	return ""
 }

+func (m Message) CozeV3StringContent() string {
+	content, ok := m.Content.(string)
+	if ok {
+		return content
+	}
+	contentList, ok := m.Content.([]any)
+	if ok {
+		contents := make([]map[string]any, 0)
+		var contentStr string
+		for _, contentItem := range contentList {
+			contentMap, ok := contentItem.(map[string]any)
+			if !ok {
+				continue
+			}
+			switch contentMap["type"] {
+			case "text":
+				if subStr, ok := contentMap["text"].(string); ok {
+					contents = append(contents, map[string]any{
+						"type": "text",
+						"text": subStr,
+					})
+				}
+			case "image_url":
+				if subStr, ok := contentMap["image_url"].(string); ok {
+					contents = append(contents, map[string]any{
+						"type":     "image",
+						"file_url": subStr,
+					})
+				}
+			case "file":
+				if subStr, ok := contentMap["image_url"].(string); ok {
+					contents = append(contents, map[string]any{
+						"type":     "file",
+						"file_url": subStr,
+					})
+				}
+			}
+		}
+		if len(contents) > 0 {
+			b, _ := json.Marshal(contents)
+			return string(b)
+		}
+		return contentStr
+	}
+	return ""
+}
+
 func (m Message) ParseContent() []MessageContent {
 	var contentList []MessageContent
 	content, ok := m.Content.(string)
@@ -71,6 +121,15 @@ func (m Message) ParseContent() []MessageContent {
 						},
 					})
 				}
+			case ContentTypeInputFile:
+				if subObj, ok := contentMap["file"].(map[string]any); ok {
+					contentList = append(contentList, MessageContent{
+						Type: ContentTypeInputFile,
+						File: &File{
+							FileData: subObj["file_data"].(string),
+						},
+					})
+				}
 			}
 		}
 		return contentList
@@ -87,4 +146,10 @@ type MessageContent struct {
 	Type     string    `json:"type,omitempty"`
 	Text     string    `json:"text"`
 	ImageURL *ImageURL `json:"image_url,omitempty"`
+	File     *File     `json:"file,omitempty"`
+}
+
+type File struct {
+	FileData string `json:"file_data,omitempty"`
+	FileName string `json:"filename,omitempty"`
 }
--- a/relay/model/misc.go
+++ b/relay/model/misc.go
@@ -4,6 +4,14 @@ type Usage struct {
 	PromptTokens     int `json:"prompt_tokens"`
 	CompletionTokens int `json:"completion_tokens"`
 	TotalTokens      int `json:"total_tokens"`
+
+	CompletionTokensDetails *CompletionTokensDetails `json:"completion_tokens_details,omitempty"`
+}
+
+type CompletionTokensDetails struct {
+	ReasoningTokens          int `json:"reasoning_tokens"`
+	AcceptedPredictionTokens int `json:"accepted_prediction_tokens"`
+	RejectedPredictionTokens int `json:"rejected_prediction_tokens"`
 }

 type Error struct {
--- a/web/air/src/constants/channel.constants.js
+++ b/web/air/src/constants/channel.constants.js
@@ -7,7 +7,7 @@ export const CHANNEL_OPTIONS = [
  { key: 24, text: 'Google Gemini', value: 24, color: 'orange' },
  { key: 28, text: 'Mistral AI', value: 28, color: 'orange' },
  { key: 41, text: 'Novita', value: 41, color: 'purple' },
-  { key: 40, text: '字节跳动豆包', value: 40, color: 'blue' },
+  {key: 40, text: '字节火山引擎', value: 40, color: 'blue'},
  { key: 15, text: '百度文心千帆', value: 15, color: 'blue' },
  { key: 17, text: '阿里通义千问', value: 17, color: 'orange' },
  { key: 18, text: '讯飞星火认知', value: 18, color: 'blue' },
@@ -22,6 +22,7 @@ export const CHANNEL_OPTIONS = [
  { key: 31, text: '零一万物', value: 31, color: 'green' },
  { key: 32, text: '阶跃星辰', value: 32, color: 'blue' },
  { key: 34, text: 'Coze', value: 34, color: 'blue' },
+  { key: 52, text: 'CozeV3', value: 52, color: 'blue' },
  { key: 35, text: 'Cohere', value: 35, color: 'blue' },
  { key: 36, text: 'DeepSeek', value: 36, color: 'black' },
  { key: 37, text: 'Cloudflare', value: 37, color: 'orange' },
@@ -35,7 +36,7 @@ export const CHANNEL_OPTIONS = [
  { key: 8, text: '自定义渠道', value: 8, color: 'pink' },
  { key: 22, text: '知识库：FastGPT', value: 22, color: 'blue' },
  { key: 21, text: '知识库：AI Proxy', value: 21, color: 'purple' },
-  { key: 20, text: '代理：OpenRouter', value: 20, color: 'black' },
+  {key: 20, text: 'OpenRouter', value: 20, color: 'black'},
  { key: 2, text: '代理：API2D', value: 2, color: 'blue' },
  { key: 5, text: '代理：OpenAI-SB', value: 5, color: 'brown' },
  { key: 7, text: '代理：OhMyGPT', value: 7, color: 'purple' },
--- a/web/berry/src/constants/ChannelConstants.js
+++ b/web/berry/src/constants/ChannelConstants.js
@@ -49,7 +49,7 @@ export const CHANNEL_OPTIONS = {
  },
  40: {
    key: 40,
-    text: '字节跳动豆包',
+    text: '字节火山引擎',
    value: 40,
    color: 'primary'
  },
@@ -137,6 +137,12 @@ export const CHANNEL_OPTIONS = {
    value: 34,
    color: 'primary'
  },
+  52: {
+    key: 52,
+    text: 'CozeV3',
+    value: 52,
+    color: 'primary'
+  },
  35: {
    key: 35,
    text: 'Cohere',
@@ -185,7 +191,7 @@ export const CHANNEL_OPTIONS = {
    value: 45,
    color: 'primary'
  },
-  45: {
+  46: {
    key: 46,
    text: 'Replicate',
    value: 46,
@@ -217,7 +223,7 @@ export const CHANNEL_OPTIONS = {
  },
  20: {
    key: 20,
-    text: '代理：OpenRouter',
+      text: 'OpenRouter',
    value: 20,
    color: 'success'
  },
--- a/web/berry/src/views/Channel/type/Config.js
+++ b/web/berry/src/views/Channel/type/Config.js
@@ -206,6 +206,20 @@ const typeConfig = {
    },
    modelGroup: 'Coze'
  },
+  52: {
+    inputLabel: {
+      config: {
+        user_id: 'User ID'
+      }
+    },
+    prompt: {
+      models: '对于 CozeV3 而言，模型名称即 Bot ID，你可以添加一个前缀 `bot-`，例如：`bot-123456`',
+      config: {
+        user_id: '生成该密钥的用户 ID'
+      }
+    },
+    modelGroup: 'CozeV3'
+  },
  42: {
    inputLabel: {
      key: '',
--- a/web/default/src/components/ChannelsTable.js
+++ b/web/default/src/components/ChannelsTable.js
@@ -1,17 +1,7 @@
-import React, { useEffect, useState } from 'react';
-import { useTranslation } from 'react-i18next';
-import {
-  Button,
-  Dropdown,
-  Form,
-  Input,
-  Label,
-  Message,
-  Pagination,
-  Popup,
-  Table,
-} from 'semantic-ui-react';
-import { Link } from 'react-router-dom';
+import React, {useEffect, useState} from 'react';
+import {useTranslation} from 'react-i18next';
+import {Button, Dropdown, Form, Input, Label, Message, Pagination, Popup, Table,} from 'semantic-ui-react';
+import {Link} from 'react-router-dom';
 import {
  API,
  loadChannelModels,
@@ -23,8 +13,8 @@ import {
  timestamp2string,
 } from '../helpers';

-import { CHANNEL_OPTIONS, ITEMS_PER_PAGE } from '../constants';
-import { renderGroup, renderNumber } from '../helpers/render';
+import {CHANNEL_OPTIONS, ITEMS_PER_PAGE} from '../constants';
+import {renderGroup, renderNumber} from '../helpers/render';

 function renderTimestamp(timestamp) {
  return <>{timestamp2string(timestamp)}</>;
@@ -54,6 +44,9 @@ function renderType(type, t) {
 function renderBalance(type, balance, t) {
  switch (type) {
    case 1: // OpenAI
+        if (balance === 0) {
+            return <span>{t('channel.table.balance_not_supported')}</span>;
+        }
      return <span>${balance.toFixed(2)}</span>;
    case 4: // CloseAI
      return <span>¥{balance.toFixed(2)}</span>;
@@ -67,6 +60,8 @@ function renderBalance(type, balance, t) {
      return <span>¥{balance.toFixed(2)}</span>;
    case 13: // AIGC2D
      return <span>{renderNumber(balance)}</span>;
+    case 20: // OpenRouter
+      return <span>${balance.toFixed(2)}</span>;
    case 36: // DeepSeek
      return <span>¥{balance.toFixed(2)}</span>;
    case 44: // SiliconFlow
@@ -93,30 +88,32 @@ const ChannelsTable = () => {
  const [showPrompt, setShowPrompt] = useState(shouldShowPrompt(promptID));
  const [showDetail, setShowDetail] = useState(isShowDetail());

+  const processChannelData = (channel) => {
+    if (channel.models === '') {
+      channel.models = [];
+      channel.test_model = '';
+    } else {
+      channel.models = channel.models.split(',');
+      if (channel.models.length > 0) {
+        channel.test_model = channel.models[0];
+      }
+      channel.model_options = channel.models.map((model) => {
+        return {
+          key: model,
+          text: model,
+          value: model,
+        };
+      });
+      console.log('channel', channel);
+    }
+    return channel;
+  };
+
  const loadChannels = async (startIdx) => {
    const res = await API.get(`/api/channel/?p=${startIdx}`);
    const { success, message, data } = res.data;
    if (success) {
-      let localChannels = data.map((channel) => {
-        if (channel.models === '') {
-          channel.models = [];
-          channel.test_model = '';
-        } else {
-          channel.models = channel.models.split(',');
-          if (channel.models.length > 0) {
-            channel.test_model = channel.models[0];
-          }
-          channel.model_options = channel.models.map((model) => {
-            return {
-              key: model,
-              text: model,
-              value: model,
-            };
-          });
-          console.log('channel', channel);
-        }
-        return channel;
-      });
+      let localChannels = data.map(processChannelData);
      if (startIdx === 0) {
        setChannels(localChannels);
      } else {
@@ -301,7 +298,8 @@ const ChannelsTable = () => {
    const res = await API.get(`/api/channel/search?keyword=${searchKeyword}`);
    const { success, message, data } = res.data;
    if (success) {
-      setChannels(data);
+      let localChannels = data.map(processChannelData);
+      setChannels(localChannels);
      setActivePage(1);
    } else {
      showError(message);
@@ -495,7 +493,6 @@ const ChannelsTable = () => {
              onClick={() => {
                sortChannel('balance');
              }}
-              hidden={!showDetail}
            >
              {t('channel.table.balance')}
            </Table.HeaderCell>
@@ -504,6 +501,7 @@ const ChannelsTable = () => {
              onClick={() => {
                sortChannel('priority');
              }}
+              hidden={!showDetail}
            >
              {t('channel.table.priority')}
            </Table.HeaderCell>
@@ -543,7 +541,7 @@ const ChannelsTable = () => {
                      basic
                    />
                  </Table.Cell>
-                  <Table.Cell hidden={!showDetail}>
+                  <Table.Cell>
                    <Popup
                      trigger={
                        <span
@@ -559,7 +557,7 @@ const ChannelsTable = () => {
                      basic
                    />
                  </Table.Cell>
-                  <Table.Cell>
+                  <Table.Cell hidden={!showDetail}>
                    <Popup
                      trigger={
                        <Input
@@ -593,7 +591,15 @@ const ChannelsTable = () => {
                    />
                  </Table.Cell>
                  <Table.Cell>
-                    <div>
+                    <div
+                      style={{
+                        display: 'flex',
+                        alignItems: 'center',
+                        flexWrap: 'wrap',
+                        gap: '2px',
+                        rowGap: '6px',
+                      }}
+                    >
                      <Button
                        size={'tiny'}
                        positive
--- a/web/default/src/constants/channel.constants.js
+++ b/web/default/src/constants/channel.constants.js
@@ -1,48 +1,109 @@
 export const CHANNEL_OPTIONS = [
-    { key: 1, text: 'OpenAI', value: 1, color: 'green' },
-    { key: 14, text: 'Anthropic Claude', value: 14, color: 'black' },
-    { key: 33, text: 'AWS', value: 33, color: 'black' },
-    { key: 3, text: 'Azure OpenAI', value: 3, color: 'olive' },
-    { key: 11, text: 'Google PaLM2', value: 11, color: 'orange' },
-    { key: 24, text: 'Google Gemini', value: 24, color: 'orange' },
-    { key: 28, text: 'Mistral AI', value: 28, color: 'orange' },
-    { key: 41, text: 'Novita', value: 41, color: 'purple' },
-    { key: 40, text: '字节跳动豆包', value: 40, color: 'blue' },
-    { key: 15, text: '百度文心千帆', value: 15, color: 'blue' },
-    { key: 17, text: '阿里通义千问', value: 17, color: 'orange' },
-    { key: 18, text: '讯飞星火认知', value: 18, color: 'blue' },
-    { key: 16, text: '智谱 ChatGLM', value: 16, color: 'violet' },
-    { key: 19, text: '360 智脑', value: 19, color: 'blue' },
-    { key: 25, text: 'Moonshot AI', value: 25, color: 'black' },
-    { key: 23, text: '腾讯混元', value: 23, color: 'teal' },
-    { key: 26, text: '百川大模型', value: 26, color: 'orange' },
-    { key: 27, text: 'MiniMax', value: 27, color: 'red' },
-    { key: 29, text: 'Groq', value: 29, color: 'orange' },
-    { key: 30, text: 'Ollama', value: 30, color: 'black' },
-    { key: 31, text: '零一万物', value: 31, color: 'green' },
-    { key: 32, text: '阶跃星辰', value: 32, color: 'blue' },
-    { key: 34, text: 'Coze', value: 34, color: 'blue' },
-    { key: 35, text: 'Cohere', value: 35, color: 'blue' },
-    { key: 36, text: 'DeepSeek', value: 36, color: 'black' },
-    { key: 37, text: 'Cloudflare', value: 37, color: 'orange' },
-    { key: 38, text: 'DeepL', value: 38, color: 'black' },
-    { key: 39, text: 'together.ai', value: 39, color: 'blue' },
-    { key: 42, text: 'VertexAI', value: 42, color: 'blue' },
-    { key: 43, text: 'Proxy', value: 43, color: 'blue' },
-    { key: 44, text: 'SiliconFlow', value: 44, color: 'blue' },
-    { key: 45, text: 'xAI', value: 45, color: 'blue' },
-    { key: 46, text: 'Replicate', value: 46, color: 'blue' },
-    { key: 8, text: '自定义渠道', value: 8, color: 'pink' },
-    { key: 22, text: '知识库：FastGPT', value: 22, color: 'blue' },
-    { key: 21, text: '知识库：AI Proxy', value: 21, color: 'purple' },
-    { key: 20, text: '代理：OpenRouter', value: 20, color: 'black' },
-    { key: 2, text: '代理：API2D', value: 2, color: 'blue' },
-    { key: 5, text: '代理：OpenAI-SB', value: 5, color: 'brown' },
-    { key: 7, text: '代理：OhMyGPT', value: 7, color: 'purple' },
-    { key: 10, text: '代理：AI Proxy', value: 10, color: 'purple' },
-    { key: 4, text: '代理：CloseAI', value: 4, color: 'teal' },
-    { key: 6, text: '代理：OpenAI Max', value: 6, color: 'violet' },
-    { key: 9, text: '代理：AI.LS', value: 9, color: 'yellow' },
-    { key: 12, text: '代理：API2GPT', value: 12, color: 'blue' },
-    { key: 13, text: '代理：AIGC2D', value: 13, color: 'purple' }
+    {key: 1, text: 'OpenAI', value: 1, color: 'green'},
+    {
+        key: 50,
+        text: 'OpenAI 兼容',
+        value: 50,
+        color: 'olive',
+        description: 'OpenAI 兼容渠道，支持设置 Base URL',
+    },
+    {key: 14, text: 'Anthropic', value: 14, color: 'black'},
+    {key: 33, text: 'AWS', value: 33, color: 'black'},
+    {key: 3, text: 'Azure', value: 3, color: 'olive'},
+    {key: 11, text: 'PaLM2', value: 11, color: 'orange'},
+    {key: 24, text: 'Gemini', value: 24, color: 'orange'},
+    {
+        key: 51,
+        text: 'Gemini (OpenAI)',
+        value: 51,
+        color: 'orange',
+        description: 'Gemini OpenAI 兼容格式',
+    },
+    {key: 28, text: 'Mistral AI', value: 28, color: 'orange'},
+    {key: 41, text: 'Novita', value: 41, color: 'purple'},
+    {
+        key: 40,
+        text: '字节火山引擎',
+        value: 40,
+        color: 'blue',
+        description: '原字节跳动豆包',
+    },
+    {
+        key: 15,
+        text: '百度文心千帆',
+        value: 15,
+        color: 'blue',
+        tip: '请前往<a href="https://console.bce.baidu.com/qianfan/ais/console/applicationConsole/application/v1" target="_blank">此处</a>获取 AK（API Key）以及 SK（Secret Key），注意，V2 版本接口请使用 <strong>百度文心千帆 V2 </strong>渠道类型',
+    },
+    {
+        key: 47,
+        text: '百度文心千帆 V2',
+        value: 47,
+        color: 'blue',
+        tip: '请前往<a href="https://console.bce.baidu.com/iam/#/iam/apikey/list" target="_blank">此处</a>获取 API Key，注意本渠道仅支持<a target="_blank" href="https://cloud.baidu.com/doc/WENXINWORKSHOP/s/em4tsqo3v">推理服务 V2</a>相关模型',
+    },
+    {
+        key: 17,
+        text: '阿里通义千问',
+        value: 17,
+        color: 'orange',
+        tip: '如需使用阿里云百炼，请使用<strong>阿里云百炼</strong>渠道',
+    },
+    {key: 49, text: '阿里云百炼', value: 49, color: 'orange'},
+    {
+        key: 18,
+        text: '讯飞星火认知',
+        value: 18,
+        color: 'blue',
+        tip: '本渠道基于讯飞 WebSocket 版本 API，如需 HTTP 版本，请使用<strong>讯飞星火认知 V2</strong>渠道',
+    },
+    {
+        key: 48,
+        text: '讯飞星火认知 V2',
+        value: 48,
+        color: 'blue',
+        tip: 'HTTP 版本的讯飞接口，前往<a href="https://console.xfyun.cn/services/cbm" target="_blank">此处</a>获取 HTTP 服务接口认证密钥',
+    },
+    {key: 16, text: '智谱 ChatGLM', value: 16, color: 'violet'},
+    {key: 19, text: '360 智脑', value: 19, color: 'blue'},
+    {key: 25, text: 'Moonshot AI', value: 25, color: 'black'},
+    {key: 23, text: '腾讯混元', value: 23, color: 'teal'},
+    {key: 26, text: '百川大模型', value: 26, color: 'orange'},
+    {key: 27, text: 'MiniMax', value: 27, color: 'red'},
+    {key: 29, text: 'Groq', value: 29, color: 'orange'},
+    {key: 30, text: 'Ollama', value: 30, color: 'black'},
+    {key: 31, text: '零一万物', value: 31, color: 'green'},
+    {key: 32, text: '阶跃星辰', value: 32, color: 'blue'},
+    {key: 34, text: 'Coze', value: 34, color: 'blue'},
+    {key: 52, text: 'CozeV3', value: 52, color: 'blue'},
+    {key: 35, text: 'Cohere', value: 35, color: 'blue'},
+    {key: 36, text: 'DeepSeek', value: 36, color: 'black'},
+    {key: 37, text: 'Cloudflare', value: 37, color: 'orange'},
+    {key: 38, text: 'DeepL', value: 38, color: 'black'},
+    {key: 39, text: 'together.ai', value: 39, color: 'blue'},
+    {key: 42, text: 'VertexAI', value: 42, color: 'blue'},
+    {key: 43, text: 'Proxy', value: 43, color: 'blue'},
+    {key: 44, text: 'SiliconFlow', value: 44, color: 'blue'},
+    {key: 45, text: 'xAI', value: 45, color: 'blue'},
+    {key: 46, text: 'Replicate', value: 46, color: 'blue'},
+    {
+        key: 8,
+        text: '自定义渠道',
+        value: 8,
+        color: 'pink',
+        tip: '不推荐使用，请使用 <strong>OpenAI 兼容</strong>渠道类型。注意，这里所需要填入的代理地址仅会在实际请求时替换域名部分，如果你想填入 OpenAI SDK 中所要求的 Base URL，请使用 OpenAI 兼容渠道类型',
+        description: '不推荐使用，请使用 OpenAI 兼容渠道类型',
+    },
+    {key: 22, text: '知识库：FastGPT', value: 22, color: 'blue'},
+    {key: 21, text: '知识库：AI Proxy', value: 21, color: 'purple'},
+    {key: 20, text: 'OpenRouter', value: 20, color: 'black'},
+    {key: 2, text: '代理：API2D', value: 2, color: 'blue'},
+    {key: 5, text: '代理：OpenAI-SB', value: 5, color: 'brown'},
+    {key: 7, text: '代理：OhMyGPT', value: 7, color: 'purple'},
+    {key: 10, text: '代理：AI Proxy', value: 10, color: 'purple'},
+    {key: 4, text: '代理：CloseAI', value: 4, color: 'teal'},
+    {key: 6, text: '代理：OpenAI Max', value: 6, color: 'violet'},
+    {key: 9, text: '代理：AI.LS', value: 9, color: 'yellow'},
+    {key: 12, text: '代理：API2GPT', value: 12, color: 'blue'},
+    {key: 13, text: '代理：AIGC2D', value: 13, color: 'purple'},
 ];
--- a/web/default/src/helpers/helper.js
+++ b/web/default/src/helpers/helper.js
@@ -0,0 +1,13 @@
+import {CHANNEL_OPTIONS} from '../constants';
+
+let channelMap = undefined;
+
+export function getChannelOption(channelId) {
+    if (channelMap === undefined) {
+        channelMap = {};
+        CHANNEL_OPTIONS.forEach((option) => {
+            channelMap[option.key] = option;
+        });
+    }
+    return channelMap[channelId];
+}
--- a/web/default/src/helpers/render.js
+++ b/web/default/src/helpers/render.js
@@ -1,5 +1,6 @@
-import { Label } from 'semantic-ui-react';
-import { useTranslation } from 'react-i18next';
+import { Label, Message } from 'semantic-ui-react';
+import { getChannelOption } from './helper';
+import React from 'react';

 export function renderText(text, limit) {
  if (text.length > limit) {
@@ -15,7 +16,15 @@ export function renderGroup(group) {
  let groups = group.split(',');
  groups.sort();
  return (
-    <>
+    <div
+      style={{
+        display: 'flex',
+        alignItems: 'center',
+        flexWrap: 'wrap',
+        gap: '2px',
+        rowGap: '6px',
+      }}
+    >
      {groups.map((group) => {
        if (group === 'vip' || group === 'pro') {
          return <Label color='yellow'>{group}</Label>;
@@ -24,7 +33,7 @@ export function renderGroup(group) {
        }
        return <Label>{group}</Label>;
      })}
-    </>
+    </div>
  );
 }

@@ -98,3 +107,15 @@ export function renderColorLabel(text) {
    </Label>
  );
 }
+
+export function renderChannelTip(channelId) {
+  let channel = getChannelOption(channelId);
+  if (channel === undefined || channel.tip === undefined) {
+    return <></>;
+  }
+  return (
+    <Message>
+      <div dangerouslySetInnerHTML={{ __html: channel.tip }}></div>
+    </Message>
+  );
+}
--- a/web/default/src/helpers/utils.js
+++ b/web/default/src/helpers/utils.js
@@ -1,7 +1,7 @@
-import { toast } from 'react-toastify';
-import { toastConstants } from '../constants';
+import {toast} from 'react-toastify';
+import {toastConstants} from '../constants';
 import React from 'react';
-import { API } from './api';
+import {API} from './api';

 const HTMLToastContent = ({ htmlContent }) => {
  return <div dangerouslySetInnerHTML={{ __html: htmlContent }} />;
@@ -74,6 +74,7 @@ if (isMobile()) {
 }

 export function showError(error) {
+  if (!error) return;
  console.error(error);
  if (error.message) {
    if (error.name === 'AxiosError') {
@@ -158,17 +159,7 @@ export function timestamp2string(timestamp) {
    second = '0' + second;
  }
  return (
-    year +
-    '-' +
-    month +
-    '-' +
-    day +
-    ' ' +
-    hour +
-    ':' +
-    minute +
-    ':' +
-    second
+      year + '-' + month + '-' + day + ' ' + hour + ':' + minute + ':' + second
  );
 }

@@ -193,7 +184,6 @@ export const verifyJSON = (str) => {
 export function shouldShowPrompt(id) {
  let prompt = localStorage.getItem(`prompt-${id}`);
  return !prompt;
-
 }

 export function setPromptShown(id) {
@@ -224,4 +214,4 @@ export function getChannelModels(type) {
    return channelModels[type];
  }
  return [];
-}
+}
--- a/web/default/src/locales/en/translation.json
+++ b/web/default/src/locales/en/translation.json
@@ -104,8 +104,10 @@
      "model_mapping_placeholder": "Optional, used to modify model names in request body. A JSON string where keys are request model names and values are target model names",
      "system_prompt": "System Prompt",
      "system_prompt_placeholder": "Optional, used to force set system prompt. Use with custom model & model mapping. First create a unique custom model name above, then map it to a natively supported model",
-      "base_url": "Proxy",
-      "base_url_placeholder": "Optional, used for API calls through proxy. Enter proxy address in format: https://domain.com",
+      "proxy_url": "Proxy",
+      "proxy_url_placeholder": "This is optional and used for API calls via a proxy. Please enter the proxy URL, formatted as: https://domain.com",
+      "base_url": "Base URL",
+      "base_url_placeholder": "The Base URL required by the OpenAPI SDK",
      "key": "Key",
      "key_placeholder": "Please enter key",
      "batch": "Batch Create",
--- a/web/default/src/locales/zh/translation.json
+++ b/web/default/src/locales/zh/translation.json
@@ -104,8 +104,10 @@
      "model_mapping_placeholder": "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",
      "system_prompt": "系统提示词",
      "system_prompt_placeholder": "此项可选，用于强制设置给定的系统提示词，请配合自定义模型 & 模型重定向使用，首先创建一个唯一的自定义模型名称并在上面填入，之后将该自定义模型重定向映射到该渠道一个原生支持的模型",
-      "base_url": "代理",
-      "base_url_placeholder": "此项可选，用于通过代理站来进行 API 调用，请输入代理站地址，格式为：https://domain.com",
+      "proxy_url": "代理",
+      "proxy_url_placeholder": "此项可选，用于通过代理站来进行 API 调用，请输入代理站地址，格式为：https://domain.com。注意，这里所需要填入的代理地址仅会在实际请求时替换域名部分，如果你想填入 OpenAI SDK 中所要求的 Base URL，请使用 OpenAI 兼容渠道类型",
+      "base_url": "Base URL",
+      "base_url_placeholder": "OpenAPI SDK 中所要求的 Base URL",
      "key": "密钥",
      "key_placeholder": "请输入密钥",
      "batch": "批量创建",
--- a/web/default/src/pages/Channel/EditChannel.js
+++ b/web/default/src/pages/Channel/EditChannel.js
@@ -1,25 +1,10 @@
-import React, { useEffect, useState } from 'react';
-import { useTranslation } from 'react-i18next';
-import {
-  Button,
-  Form,
-  Header,
-  Input,
-  Message,
-  Segment,
-  Card,
-} from 'semantic-ui-react';
-import { useNavigate, useParams } from 'react-router-dom';
-import {
-  API,
-  copy,
-  getChannelModels,
-  showError,
-  showInfo,
-  showSuccess,
-  verifyJSON,
-} from '../../helpers';
-import { CHANNEL_OPTIONS } from '../../constants';
+import React, {useEffect, useState} from 'react';
+import {useTranslation} from 'react-i18next';
+import {Button, Card, Form, Input, Message} from 'semantic-ui-react';
+import {useNavigate, useParams} from 'react-router-dom';
+import {API, copy, getChannelModels, showError, showInfo, showSuccess, verifyJSON,} from '../../helpers';
+import {CHANNEL_OPTIONS} from '../../constants';
+import {renderChannelTip} from '../../helpers/render';

 const MODEL_MAPPING_EXAMPLE = {
  'gpt-3.5-turbo-0301': 'gpt-3.5-turbo',
@@ -310,6 +295,7 @@ const EditChannel = () => {
                options={groupOptions}
              />
            </Form.Field>
+            {renderChannelTip(inputs.type)}

            {/* Azure OpenAI specific fields */}
            {inputs.type === 3 && (
@@ -353,6 +339,20 @@ const EditChannel = () => {
            {inputs.type === 8 && (
              <Form.Field>
                <Form.Input
+                    required
+                    label={t('channel.edit.proxy_url')}
+                    name='base_url'
+                    placeholder={t('channel.edit.proxy_url_placeholder')}
+                    onChange={handleInputChange}
+                    value={inputs.base_url}
+                    autoComplete='new-password'
+                />
+              </Form.Field>
+            )}
+            {inputs.type === 50 && (
+                <Form.Field>
+                  <Form.Input
+                      required
                  label={t('channel.edit.base_url')}
                  name='base_url'
                  placeholder={t('channel.edit.base_url_placeholder')}
@@ -651,12 +651,13 @@ const EditChannel = () => {
            {inputs.type !== 3 &&
              inputs.type !== 33 &&
              inputs.type !== 8 &&
+                inputs.type !== 50 &&
              inputs.type !== 22 && (
                <Form.Field>
                  <Form.Input
-                    label={t('channel.edit.base_url')}
+                      label={t('channel.edit.proxy_url')}
                    name='base_url'
-                    placeholder={t('channel.edit.base_url_placeholder')}
+                      placeholder={t('channel.edit.proxy_url_placeholder')}
                    onChange={handleInputChange}
                    value={inputs.base_url}
                    autoComplete='new-password'
--- a/web/default/src/pages/Dashboard/index.js
+++ b/web/default/src/pages/Dashboard/index.js
@@ -122,11 +122,11 @@ const Dashboard = () => {
        ? new Date(Math.min(...dates.map((d) => new Date(d))))
        : new Date();

-    // 确保至少显示5天的数据
-    const fiveDaysAgo = new Date();
-    fiveDaysAgo.setDate(fiveDaysAgo.getDate() - 4); // -4是因为包含今天
-    if (minDate > fiveDaysAgo) {
-      minDate = fiveDaysAgo;
+    // 确保至少显示7天的数据
+    const sevenDaysAgo = new Date();
+    sevenDaysAgo.setDate(sevenDaysAgo.getDate() - 6); // -6是因为包含今天
+    if (minDate > sevenDaysAgo) {
+      minDate = sevenDaysAgo;
    }

    // 生成所有日期
@@ -164,11 +164,11 @@ const Dashboard = () => {
        ? new Date(Math.min(...dates.map((d) => new Date(d))))
        : new Date();

-    // 确保至少显示5天的数据
-    const fiveDaysAgo = new Date();
-    fiveDaysAgo.setDate(fiveDaysAgo.getDate() - 4); // -4是因为包含今天
-    if (minDate > fiveDaysAgo) {
-      minDate = fiveDaysAgo;
+    // 确保至少显示7天的数据
+    const sevenDaysAgo = new Date();
+    sevenDaysAgo.setDate(sevenDaysAgo.getDate() - 6); // -6是因为包含今天
+    if (minDate > sevenDaysAgo) {
+      minDate = sevenDaysAgo;
    }

    // 生成所有日期
@@ -242,7 +242,7 @@ const Dashboard = () => {
            <Card.Content>
              <Card.Header>
                {t('dashboard.charts.requests.title')}
-                  {/* <span className='stat-value'>{summaryData.todayRequests}</span> */}
+                {/* <span className='stat-value'>{summaryData.todayRequests}</span> */}
              </Card.Header>
              <div className='chart-container'>
                <ResponsiveContainer
@@ -271,7 +271,9 @@ const Dashboard = () => {
                        t('dashboard.charts.requests.tooltip'),
                      ]}
                      labelFormatter={(label) =>
-                        `${t('dashboard.statistics.tooltip.date')}: ${formatDate(label)}`
+                        `${t(
+                          'dashboard.statistics.tooltip.date'
+                        )}: ${formatDate(label)}`
                      }
                    />
                    <Line
@@ -294,7 +296,7 @@ const Dashboard = () => {
            <Card.Content>
              <Card.Header>
                {t('dashboard.charts.quota.title')}
-                  {/* <span className='stat-value'>
+                {/* <span className='stat-value'>
                  ${summaryData.todayQuota.toFixed(3)}
                </span> */}
              </Card.Header>
@@ -321,11 +323,13 @@ const Dashboard = () => {
                        boxShadow: '0 2px 8px rgba(0,0,0,0.1)',
                      }}
                      formatter={(value) => [
-                        value,
+                        value.toFixed(6),
                        t('dashboard.charts.quota.tooltip'),
                      ]}
                      labelFormatter={(label) =>
-                        `${t('dashboard.statistics.tooltip.date')}: ${formatDate(label)}`
+                        `${t(
+                          'dashboard.statistics.tooltip.date'
+                        )}: ${formatDate(label)}`
                      }
                    />
                    <Line
@@ -348,7 +352,7 @@ const Dashboard = () => {
            <Card.Content>
              <Card.Header>
                {t('dashboard.charts.tokens.title')}
-                  {/* <span className='stat-value'>{summaryData.todayTokens}</span> */}
+                {/* <span className='stat-value'>{summaryData.todayTokens}</span> */}
              </Card.Header>
              <div className='chart-container'>
                <ResponsiveContainer
@@ -377,7 +381,9 @@ const Dashboard = () => {
                        t('dashboard.charts.tokens.tooltip'),
                      ]}
                      labelFormatter={(label) =>
-                        `${t('dashboard.statistics.tooltip.date')}: ${formatDate(label)}`
+                        `${t(
+                          'dashboard.statistics.tooltip.date'
+                        )}: ${formatDate(label)}`
                      }
                    />
                    <Line
@@ -422,7 +428,9 @@ const Dashboard = () => {
                    boxShadow: '0 2px 8px rgba(0,0,0,0.1)',
                  }}
                  labelFormatter={(label) =>
-                    `${t('dashboard.statistics.tooltip.date')}: ${formatDate(label)}`
+                    `${t('dashboard.statistics.tooltip.date')}: ${formatDate(
+                      label
+                    )}`
                  }
                />
                <Legend
Author	SHA1	Message	Date
suziheng	974331a028	feat:gemini file	2025-09-15 17:16:50 +08:00
suziheng	a529eab39e	build:dockerfile	2025-09-15 16:18:27 +08:00
suziheng	2174039fce	build:dockerfile	2025-09-15 15:58:46 +08:00
suziheng	9ce714ac8d	build:dockerfile	2025-09-15 15:33:48 +08:00
suziheng	7d2fc27c0f	build:dockerfile	2025-09-15 15:30:56 +08:00
suziheng	4d3add220e	build: npm version	2025-09-15 15:17:55 +08:00
suziheng	04de01c798	feat: support openai format enable deepseek v3.1 thinking	2025-09-15 14:43:19 +08:00
suziheng	cf0ce425e6	feat: 调整gemini tools	2025-05-28 15:08:48 +08:00
suziheng	e1ee4fe7d9	feat: 支持gemini 思考过程	2025-05-21 16:19:36 +08:00
suziheng	1e19c333c9	fix: get gemini adapter bug	2025-05-16 10:18:07 +08:00
suziheng	93d54a7ef5	feat: 增加gpt-4.1	2025-04-23 10:27:45 +08:00
suziheng	9a7967e9bb	feat: 调整azure deployment	2025-04-23 10:25:35 +08:00
suziheng	c7742de0fc	build: 并行改为串行构建	2025-04-22 11:52:57 +08:00
suziheng	a8a303b4ee	feat: 恢复表结构	2025-04-22 11:12:23 +08:00
suziheng	abf9d113af	feat: 恢复表结构	2025-04-22 11:10:21 +08:00
suziheng	5f5521bc9a	feat: 更新模型倍率	2025-04-22 09:27:58 +08:00
suziheng	77267aa1b8	feat: 合并main	2025-04-21 16:36:54 +08:00
suziheng	dfcf8868fe	Merge branch 'feat/coze-v3' into feat/transcribe	2025-04-21 16:03:35 +08:00
suziheng	c2bd301e0a	feat:修复transcribe bug	2025-04-21 15:59:44 +08:00
JustSong	8df4a2670b	docs: update ByteDance Doubao model link in README Some checks failed CI / Unit tests (push) Has been cancelled Details CI / commit_lint (push) Has been cancelled Details	2025-02-21 19:30:16 +08:00
longkeyy	7ac553541b	feat: update openrouter models and price 20250213 (#2084 ) Some checks failed CI / Unit tests (push) Has been cancelled Details CI / commit_lint (push) Has been cancelled Details	2025-02-16 18:01:59 +08:00
longkeyy	a5c517c27a	feat: update ali models and price 20250213 (#2086 )	2025-02-16 18:01:24 +08:00
JustSong	3f421c4f04	feat: support Gemini openai compatible api	2025-02-16 17:59:39 +08:00
JustSong	1ce6a226f6	chore: update prompt	2025-02-16 17:42:20 +08:00
JustSong	cafd0a0327	feat: add OpenAI compatible channel (close #2091 )	2025-02-16 17:38:06 +08:00
JustSong	8b8cd03e85	feat: add balance not supported message in ChannelsTable Some checks failed CI / Unit tests (push) Has been cancelled Details CI / commit_lint (push) Has been cancelled Details	2025-02-12 01:20:28 +08:00
JustSong	54c38de813	style: improve code formatting and structure in ChannelsTable and render helpers	2025-02-12 01:15:45 +08:00
JustSong	d6284bf6b0	feat: enhance error handling for utils.js	2025-02-12 00:46:13 +08:00
DobyAsa	df5d2ca93d	docs: fix README typo (#2060 )	2025-02-12 00:35:29 +08:00
Laisky.Cai	fef7ae048b	feat: support gemini-2.0-flash (#2055 ) * feat: support gemini-2.0-flash - Enhance model support by adding new entries and refining checks for system instruction compatibility. - Update logging display behavior and adjust default quotas for better user experience. - Revamp pricing structures in the billing system to reflect current model values and deprecate outdated entries. - Streamline code by replacing hardcoded values with configurations for maintainability. * feat: add new Gemini 2.0 flash models to adapter and billing ratio * fix: update GetRequestURL to support gemini-1.5 model in versioning	2025-02-12 00:34:25 +08:00
JustSong	6916debf66	feat: update TestPrompt to specify output format for model name	2025-02-12 00:28:23 +08:00
JustSong	53da209134	feat: add AliBailian adaptor and update channel options	2025-02-12 00:15:43 +08:00
JustSong	517f6ad211	feat: update date range to display at least 7 days of data in Dashboard Some checks are pending CI / Unit tests (push) Waiting to run Details CI / commit_lint (push) Waiting to run Details	2025-02-11 01:48:26 +08:00
JustSong	10aba11f18	style: improve code formatting in ChannelsTable component	2025-02-11 00:38:15 +08:00
JustSong	4d011c5f98	feat: add OpenRouter balance update functionality and improve code formatting	2025-02-11 00:35:06 +08:00
JustSong	eb96aa635e	feat: update OpenRouter channel name and add model list for OpenRouter adaptor	2025-02-11 00:20:55 +08:00
JustSong	c715f2bc1d	feat: add new models for xai Some checks are pending CI / Unit tests (push) Waiting to run Details CI / commit_lint (push) Waiting to run Details	2025-02-09 21:21:28 +08:00
JustSong	aed090dd55	fix: fix cannot select test model when searching Some checks are pending CI / Unit tests (push) Waiting to run Details CI / commit_lint (push) Waiting to run Details	2025-02-09 19:09:53 +08:00
JustSong	696265774e	feat: add MiniMax model constants to the adaptor	2025-02-09 18:55:32 +08:00
JustSong	974729426d	feat: refactor Xunfei API version handling and update model list	2025-02-09 18:50:51 +08:00
JustSong	57c1367ec8	feat: add Xunfei V2 channel support and update related configurations	2025-02-09 18:31:54 +08:00
JustSong	44233d5c04	feat: add completion tokens details and reasoning effort fields to model (close #2050 )	2025-02-09 18:14:01 +08:00
JustSong	bf45a955c3	fix: update system prompt handling by renaming field and ensuring proper usage in request processing (close #2069 )	2025-02-09 14:41:42 +08:00
JustSong	20435fcbfc	fix: simplify Docker build configuration by removing unnecessary platform and architecture settings	2025-02-09 14:33:25 +08:00
JustSong	6e7a1c2323	fix: format channel options for consistency and improve tips for user guidance	2025-02-09 12:42:31 +08:00
JustSong	dd65b997dd	feat: add Baidu V2 channel support and improve model handling	2025-02-09 12:37:26 +08:00
JustSong	0b6d03d6c6	fix: update channel name from '火山引擎' to '字节火山引擎' for consistency	2025-02-09 12:08:40 +08:00
JustSong	4375246e24	feat: enhance channel options with tips and descriptions for better user guidance	2025-02-09 12:03:31 +08:00
longkeyy	3e3b8230ac	fix: add read/write locks for ModelRatio and GroupRatio to prevent concurrent map read/write issues (#2067 ) Some checks are pending CI / Unit tests (push) Waiting to run Details CI / commit_lint (push) Waiting to run Details	2025-02-09 11:02:45 +08:00
JustSong	07808122a6	fix: fix Debugf not using DebugEnabled (close #2068 )	2025-02-09 10:57:22 +08:00
牡丹凤凰	c96895e35b	docs: add related project CherryStudio (#2059 ) Some checks failed CI / Unit tests (push) Has been cancelled Details CI / commit_lint (push) Has been cancelled Details * Update README.md 增加相关项目CherryStudio * Update README.en.md * Update README.ja.md	2025-02-08 00:07:55 +08:00
JustSong	2552c68249	fix: update doubao channel name Some checks are pending CI / Unit tests (push) Waiting to run Details CI / commit_lint (push) Waiting to run Details	2025-02-07 01:51:28 +08:00
JustSong	5c81e40612	fix: update Dockerfile and workflow for improved multi-architecture support	2025-02-07 01:35:53 +08:00
JustSong	0d5318b1b7	revert: fix: revert sqlite build related changes This reverts commit `db65db2807`.	2025-02-07 01:15:33 +08:00
JustSong	db65db2807	fix: revert sqlite build related changes	2025-02-07 00:48:23 +08:00
JustSong	e0b7e6a9e2	fix: unify version retrieval in Dockerfile build commands	2025-02-07 00:39:55 +08:00
JustSong	27c2abe80f	fix: update Docker setup actions to latest versions	2025-02-07 00:33:15 +08:00
JustSong	2c867251b5	fix: improve code formatting and readability in Dashboard component	2025-02-07 00:23:13 +08:00
JustSong	108111ebd3	fix: exclude preview tags from release workflows	2025-02-07 00:19:23 +08:00
JustSong	293ba93ad6	fix: remove outdated model from ModelList and add new deepseek models	2025-02-07 00:13:57 +08:00
JustSong	faced40d5b	fix: update Docker image workflow to conditionally include arm64 platform	2025-02-07 00:06:32 +08:00
suziheng	cbf8413a39	feat: 支持CozeV3	2025-01-22 21:37:17 +08:00
suziheng	dde3cff708	feat: 支持CozeV3	2025-01-22 19:49:24 +08:00
suziheng	aca72dc979	feat: 支持CozeV3	2025-01-22 18:10:25 +08:00
suziheng	533f9853ac	feat: 支持CozeV3	2025-01-22 16:43:40 +08:00
zephyrs988	9746803a2f	Merge branch 'songquanpeng:main' into main	2025-01-16 11:58:45 +08:00
suziheng	6eb4e788c7	feat: 增加迁移开关	2024-12-24 10:08:38 +08:00
suziheng	f8fcb1d258	feat: 更新代码	2024-12-24 10:03:50 +08:00
suziheng	9c931b7d43	修改dockerfile	2024-12-02 11:08:36 +08:00
suziheng	4882fd60ab	修改dockerfile	2024-12-02 11:01:30 +08:00
suziheng	a9f42abb59	设置迁移开关	2024-12-02 10:32:39 +08:00