merge upstream

Signed-off-by: wozulong <>

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

@@ -4,7 +4,6 @@ on:
   push:
     tags:
       - '*'
-      - '!*-alpha*'
   workflow_dispatch:
     inputs:
       name:

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

@@ -4,7 +4,6 @@ on:
   push:
     tags:
       - '*'
-      - '!*-alpha*'
   workflow_dispatch:
     inputs:
       name:

common/constants.go

@@ -55,6 +55,7 @@ var WeChatAuthEnabled = false
 var TelegramOAuthEnabled = false
 var TurnstileCheckEnabled = false
 var RegisterEnabled = true
+var UserSelfDeletionEnabled = false
 
 var EmailDomainRestrictionEnabled = false
 var EmailDomainWhitelist = []string{

common/model-ratio.go

@@ -73,6 +73,9 @@ var DefaultModelRatio = map[string]float64{
 	"PaLM-2":                    1,
 	"gemini-pro":                1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
 	"gemini-pro-vision":         1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
+	"gemini-1.0-pro-vision-001": 1,
+	"gemini-1.0-pro-001":        1,
+	"gemini-1.5-pro":            1,
 	"chatglm_turbo":             0.3572, // ￥0.005 / 1k tokens
 	"chatglm_pro":               0.7143, // ￥0.01 / 1k tokens
 	"chatglm_std":               0.3572, // ￥0.005 / 1k tokens
@@ -212,5 +215,15 @@ func GetCompletionRatio(name string) float64 {
 	} else if strings.HasPrefix(name, "claude-3") {
 		return 5
 	}
+	if strings.HasPrefix(name, "mistral-") {
+		return 3
+	}
+	if strings.HasPrefix(name, "gemini-") {
+		return 3
+	}
+	switch name {
+	case "llama2-70b-4096":
+		return 0.8 / 0.7
+	}
 	return 1
 }

common/redis.go

@@ -18,9 +18,8 @@ func InitRedisClient() (err error) {
 		return nil
 	}
 	if os.Getenv("SYNC_FREQUENCY") == "" {
-		RedisEnabled = false
-		SysLog("SYNC_FREQUENCY not set, Redis is disabled")
-		return nil
+		SysLog("SYNC_FREQUENCY not set, use default value 60")
+		SyncFrequency = 60
 	}
 	SysLog("Redis is enabled")
 	opt, err := redis.ParseURL(os.Getenv("REDIS_CONN_STRING"))

controller/channel-test.go

@@ -108,6 +108,7 @@ func buildTestRequest() *dto.GeneralOpenAIRequest {
 	testRequest := &dto.GeneralOpenAIRequest{
 		Model:     "", // this will be set later
 		MaxTokens: 1,
+		Stream:    false,
 	}
 	content, _ := json.Marshal("hi")
 	testMessage := dto.Message{

controller/user.go

@@ -558,6 +558,14 @@ func HardDeleteUser(c *gin.Context) {
 }
 
 func DeleteSelf(c *gin.Context) {
+	if !common.UserSelfDeletionEnabled {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": "当前设置不允许用户自我删除账号",
+		})
+		return
+	}
+
 	id := c.GetInt("id")
 	user, _ := model.GetUserById(id, false)
 

model/option.go

@@ -36,6 +36,7 @@ func InitOptionMap() {
 	common.OptionMap["WeChatAuthEnabled"] = strconv.FormatBool(common.WeChatAuthEnabled)
 	common.OptionMap["TurnstileCheckEnabled"] = strconv.FormatBool(common.TurnstileCheckEnabled)
 	common.OptionMap["RegisterEnabled"] = strconv.FormatBool(common.RegisterEnabled)
+	common.OptionMap["UserSelfDeletionEnabled"] = strconv.FormatBool(common.UserSelfDeletionEnabled)
 	common.OptionMap["AutomaticDisableChannelEnabled"] = strconv.FormatBool(common.AutomaticDisableChannelEnabled)
 	common.OptionMap["AutomaticEnableChannelEnabled"] = strconv.FormatBool(common.AutomaticEnableChannelEnabled)
 	common.OptionMap["LogConsumeEnabled"] = strconv.FormatBool(common.LogConsumeEnabled)
@@ -177,6 +178,8 @@ func updateOptionMap(key string, value string) (err error) {
 			common.TurnstileCheckEnabled = boolValue
 		case "RegisterEnabled":
 			common.RegisterEnabled = boolValue
+		case "UserSelfDeletionEnabled":
+			common.UserSelfDeletionEnabled = boolValue
 		case "EmailDomainRestrictionEnabled":
 			common.EmailDomainRestrictionEnabled = boolValue
 		case "AutomaticDisableChannelEnabled":

relay/channel/gemini/constant.go

@@ -5,8 +5,8 @@ const (
 )
 
 var ModelList = []string{
-	"gemini-pro",
-	"gemini-pro-vision",
+	"gemini-pro", "gemini-1.0-pro-001", "gemini-1.5-pro",
+	"gemini-pro-vision", "gemini-1.0-pro-vision-001",
 }
 
 var ChannelName = "google gemini"

relay/channel/ollama/adaptor.go

@@ -2,7 +2,6 @@ package ollama
 
 import (
 	"errors"
-	"fmt"
 	"github.com/gin-gonic/gin"
 	"io"
 	"net/http"
@@ -10,6 +9,7 @@ import (
 	"one-api/relay/channel"
 	"one-api/relay/channel/openai"
 	relaycommon "one-api/relay/common"
+	relayconstant "one-api/relay/constant"
 	"one-api/service"
 )
 
@@ -20,7 +20,12 @@ func (a *Adaptor) Init(info *relaycommon.RelayInfo, request dto.GeneralOpenAIReq
 }
 
 func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
-	return fmt.Sprintf("%s/api/chat", info.BaseUrl), nil
+	switch info.RelayMode {
+	case relayconstant.RelayModeEmbeddings:
+		return info.BaseUrl + "/api/embeddings", nil
+	default:
+		return relaycommon.GetFullRequestURL(info.BaseUrl, info.RequestURLPath, info.ChannelType), nil
+	}
 }
 
 func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, info *relaycommon.RelayInfo) error {
@@ -32,7 +37,12 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *dto.Gen
 	if request == nil {
 		return nil, errors.New("request is nil")
 	}
+	switch relayMode {
+	case relayconstant.RelayModeEmbeddings:
+		return requestOpenAI2Embeddings(*request), nil
+	default:
 		return requestOpenAI2Ollama(*request), nil
+	}
 }
 
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (*http.Response, error) {
@@ -44,9 +54,13 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 		var responseText string
 		err, responseText = openai.OpenaiStreamHandler(c, resp, info.RelayMode)
 		usage, _ = service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens)
+	} else {
+		if info.RelayMode == relayconstant.RelayModeEmbeddings {
+			err, usage, sensitiveResp = ollamaEmbeddingHandler(c, resp, info.PromptTokens, info.UpstreamModelName, info.RelayMode)
 		} else {
 			err, usage, sensitiveResp = openai.OpenaiHandler(c, resp, info.PromptTokens, info.UpstreamModelName, info.RelayMode)
 		}
+	}
 	return
 }
 

relay/channel/ollama/dto.go

@@ -6,13 +6,21 @@ type OllamaRequest struct {
 	Model       string        `json:"model,omitempty"`
 	Messages    []dto.Message `json:"messages,omitempty"`
 	Stream      bool          `json:"stream,omitempty"`
-	Options  *OllamaOptions `json:"options,omitempty"`
-}
-
-type OllamaOptions struct {
 	Temperature float64       `json:"temperature,omitempty"`
 	Seed        float64       `json:"seed,omitempty"`
 	Topp        float64       `json:"top_p,omitempty"`
 	TopK        int           `json:"top_k,omitempty"`
 	Stop        any           `json:"stop,omitempty"`
 }
+
+type OllamaEmbeddingRequest struct {
+	Model  string `json:"model,omitempty"`
+	Prompt any    `json:"prompt,omitempty"`
+}
+
+type OllamaEmbeddingResponse struct {
+	Embedding []float64 `json:"embedding,omitempty"`
+}
+
+//type OllamaOptions struct {
+//}

relay/channel/ollama/relay-ollama.go

@@ -1,6 +1,15 @@
 package ollama
 
-import "one-api/dto"
+import (
+	"bytes"
+	"encoding/json"
+	"fmt"
+	"github.com/gin-gonic/gin"
+	"io"
+	"net/http"
+	"one-api/dto"
+	"one-api/service"
+)
 
 func requestOpenAI2Ollama(request dto.GeneralOpenAIRequest) *OllamaRequest {
 	messages := make([]dto.Message, 0, len(request.Messages))
@@ -21,12 +30,79 @@ func requestOpenAI2Ollama(request dto.GeneralOpenAIRequest) *OllamaRequest {
 		Model:       request.Model,
 		Messages:    messages,
 		Stream:      request.Stream,
-		Options: &OllamaOptions{
 		Temperature: request.Temperature,
 		Seed:        request.Seed,
 		Topp:        request.TopP,
 		TopK:        request.TopK,
 		Stop:        Stop,
-		},
 	}
 }
+
+func requestOpenAI2Embeddings(request dto.GeneralOpenAIRequest) *OllamaEmbeddingRequest {
+	return &OllamaEmbeddingRequest{
+		Model:  request.Model,
+		Prompt: request.Input,
+	}
+}
+
+func ollamaEmbeddingHandler(c *gin.Context, resp *http.Response, promptTokens int, model string, relayMode int) (*dto.OpenAIErrorWithStatusCode, *dto.Usage, *dto.SensitiveResponse) {
+	var ollamaEmbeddingResponse OllamaEmbeddingResponse
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil, nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil, nil
+	}
+	err = json.Unmarshal(responseBody, &ollamaEmbeddingResponse)
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil, nil
+	}
+	data := make([]dto.OpenAIEmbeddingResponseItem, 0, 1)
+	data = append(data, dto.OpenAIEmbeddingResponseItem{
+		Embedding: ollamaEmbeddingResponse.Embedding,
+		Object:    "embedding",
+	})
+	usage := &dto.Usage{
+		TotalTokens:      promptTokens,
+		CompletionTokens: 0,
+		PromptTokens:     promptTokens,
+	}
+	embeddingResponse := &dto.OpenAIEmbeddingResponse{
+		Object: "list",
+		Data:   data,
+		Model:  model,
+		Usage:  *usage,
+	}
+	doResponseBody, err := json.Marshal(embeddingResponse)
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil, nil
+	}
+	resp.Body = io.NopCloser(bytes.NewBuffer(doResponseBody))
+	// We shouldn't set the header before we parse the response body, because the parse part may fail.
+	// And then we will have to send an error response, but in this case, the header has already been set.
+	// So the httpClient will be confused by the response.
+	// For example, Postman will report error, and we cannot check the response at all.
+	// Copy headers
+	for k, v := range resp.Header {
+		// 删除任何现有的相同头部，以防止重复添加头部
+		c.Writer.Header().Del(k)
+		for _, vv := range v {
+			c.Writer.Header().Add(k, vv)
+		}
+	}
+	// reset content length
+	c.Writer.Header().Del("Content-Length")
+	c.Writer.Header().Set("Content-Length", fmt.Sprintf("%d", len(doResponseBody)))
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, err = io.Copy(c.Writer, resp.Body)
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "copy_response_body_failed", http.StatusInternalServerError), nil, nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil, nil
+	}
+	return nil, usage, nil
+}

relay/channel/openai/adaptor.go

@@ -10,8 +10,8 @@ import (
 	"one-api/dto"
 	"one-api/relay/channel"
 	"one-api/relay/channel/ai360"
-	"one-api/relay/channel/moonshot"
 	"one-api/relay/channel/lingyiwanwu"
+	"one-api/relay/channel/moonshot"
 	relaycommon "one-api/relay/common"
 	"one-api/service"
 	"strings"
@@ -34,9 +34,6 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 		model_ := info.UpstreamModelName
 		model_ = strings.Replace(model_, ".", "", -1)
 		// https://github.com/songquanpeng/one-api/issues/67
-		model_ = strings.TrimSuffix(model_, "-0301")
-		model_ = strings.TrimSuffix(model_, "-0314")
-		model_ = strings.TrimSuffix(model_, "-0613")
 
 		requestURL = fmt.Sprintf("/openai/deployments/%s/%s", model_, task)
 		return relaycommon.GetFullRequestURL(info.BaseUrl, requestURL, info.ChannelType), nil

relay/relay-text.go

@@ -162,7 +162,7 @@ func TextHelper(c *gin.Context) *dto.OpenAIErrorWithStatusCode {
 
 	if resp.StatusCode != http.StatusOK {
 		returnPreConsumedQuota(c, relayInfo.TokenId, userQuota, preConsumedQuota)
-		return service.RelayErrorHandler(resp)
+		return service.OpenAIErrorWrapper(fmt.Errorf("bad response status code: %d", resp.StatusCode), "bad_response_status_code", resp.StatusCode)
 	}
 
 	usage, openaiErr, sensitiveResp := adaptor.DoResponse(c, resp, relayInfo)

service/error.go

@@ -29,7 +29,7 @@ func MidjourneyErrorWithStatusCodeWrapper(code int, desc string, statusCode int)
 func OpenAIErrorWrapper(err error, code string, statusCode int) *dto.OpenAIErrorWithStatusCode {
 	text := err.Error()
 	// 定义一个正则表达式匹配URL
-	if strings.Contains(text, "Post") {
+	if strings.Contains(text, "Post") || strings.Contains(text, "dial") {
 		common.SysLog(fmt.Sprintf("error: %s", text))
 		text = "请求上游地址失败"
 	}

service/token_counter.go

@@ -208,7 +208,7 @@ func CountTokenInput(input any, model string, check bool) (int, error, bool) {
 		}
 		return CountTokenText(text, model, check)
 	}
-	return 0, errors.New("unsupported input type"), false
+	return CountTokenInput(fmt.Sprintf("%v", input), model, check)
 }
 
 func CountAudioToken(text string, model string, check bool) (int, error, bool) {

web/src/components/SystemSetting.js

@@ -45,6 +45,7 @@ const SystemSetting = () => {
     TurnstileSiteKey: '',
     TurnstileSecretKey: '',
     RegisterEnabled: '',
+    UserSelfDeletionEnabled: false,
     EmailDomainRestrictionEnabled: '',
     EmailDomainWhitelist: [],
     // telegram login
@@ -104,6 +105,7 @@ const SystemSetting = () => {
       case 'TurnstileCheckEnabled':
       case 'EmailDomainRestrictionEnabled':
       case 'RegisterEnabled':
+      case 'UserSelfDeletionEnabled':
       case 'PaymentEnabled':
         value = inputs[key] === 'true' ? 'false' : 'true';
         break;
@@ -536,6 +538,12 @@ const SystemSetting = () => {
               name='TurnstileCheckEnabled'
               onChange={handleInputChange}
             />
+            <Form.Checkbox
+                checked={inputs.UserSelfDeletionEnabled === 'true'}
+                label='允许用户自行删除账户'
+                name='UserSelfDeletionEnabled'
+                onChange={handleInputChange}
+            />
           </Form.Group>
           <Divider />
           <Header as='h3'>

web/src/components/TokensTable.js

@@ -426,8 +426,11 @@ const TokensTable = () => {
     if (await copy(text)) {
       showSuccess('已复制到剪贴板！');
     } else {
-      // setSearchKeyword(text);
-      Modal.error({ title: '无法复制到剪贴板，请手动复制', content: text });
+      Modal.error({
+        title: '无法复制到剪贴板，请手动复制',
+        content: text,
+        size: 'large',
+      });
     }
   };