feat: support Ali stable-diffusion-xl and wanx-v1 model (#1240)

* Fix ali ConvertRequest function to use baidu keyword

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

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

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

* chore: update ali constants and model ratio

---------

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

relay/channel/openai/adaptor.go

@@ -7,6 +7,7 @@ import (
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/relay/channel"
 	"github.com/songquanpeng/one-api/relay/channel/minimax"
+	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/util"
 	"io"
@@ -25,6 +26,13 @@ func (a *Adaptor) Init(meta *util.RelayMeta) {
 func (a *Adaptor) GetRequestURL(meta *util.RelayMeta) (string, error) {
 	switch meta.ChannelType {
 	case common.ChannelTypeAzure:
+		if meta.Mode == constant.RelayModeImagesGenerations {
+			// https://learn.microsoft.com/en-us/azure/ai-services/openai/dall-e-quickstart?tabs=dalle3%2Ccommand-line&pivots=rest-api
+			// https://{resource_name}.openai.azure.com/openai/deployments/dall-e-3/images/generations?api-version=2024-03-01-preview
+			fullRequestURL := fmt.Sprintf("%s/openai/deployments/%s/images/generations?api-version=%s", meta.BaseURL, meta.ActualModelName, meta.APIVersion)
+			return fullRequestURL, nil
+		}
+
 		// https://learn.microsoft.com/en-us/azure/cognitive-services/openai/chatgpt-quickstart?pivots=rest-api&tabs=command-line#rest-api
 		requestURL := strings.Split(meta.RequestURLPath, "?")[0]
 		requestURL = fmt.Sprintf("%s?api-version=%s", requestURL, meta.APIVersion)
@@ -63,6 +71,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	return request, nil
 }
 
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
 	return channel.DoRequestHelper(a, c, meta, requestBody)
 }
@@ -75,7 +90,12 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *util.Rel
 			usage = ResponseText2Usage(responseText, meta.ActualModelName, meta.PromptTokens)
 		}
 	} else {
-		err, usage = Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
+		switch meta.Mode {
+		case constant.RelayModeImagesGenerations:
+			err, usage = ImageHandler(c, resp)
+		default:
+			err, usage = Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
+		}
 	}
 	return
 }

relay/channel/openai/main.go

@@ -149,3 +149,37 @@ func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName st
 	}
 	return nil, &textResponse.Usage
 }
+
+func ImageHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	var imageResponse ImageResponse
+	responseBody, err := io.ReadAll(resp.Body)
+
+	if err != nil {
+		return ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = json.Unmarshal(responseBody, &imageResponse)
+	if err != nil {
+		return ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+
+	resp.Body = io.NopCloser(bytes.NewBuffer(responseBody))
+
+	for k, v := range resp.Header {
+		c.Writer.Header().Set(k, v[0])
+	}
+	c.Writer.WriteHeader(resp.StatusCode)
+
+	_, err = io.Copy(c.Writer, resp.Body)
+	if err != nil {
+		return ErrorWrapper(err, "copy_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	return nil, &imageResponse.Usage
+}

relay/channel/openai/model.go

@@ -110,11 +110,16 @@ type EmbeddingResponse struct {
 	model.Usage `json:"usage"`
 }
 
+type ImageData struct {
+	Url           string `json:"url,omitempty"`
+	B64Json       string `json:"b64_json,omitempty"`
+	RevisedPrompt string `json:"revised_prompt,omitempty"`
+}
+
 type ImageResponse struct {
-	Created int `json:"created"`
-	Data    []struct {
-		Url string `json:"url"`
-	}
+	Created     int         `json:"created"`
+	Data        []ImageData `json:"data"`
+	model.Usage `json:"usage"`
 }
 
 type ChatCompletionsStreamResponseChoice struct {