xinyin025/one-api
1
0
Fork 0
mirror of https://github.com/songquanpeng/one-api.git synced 2025-11-12 03:13:41 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge branch 'upstream/main'

Browse Source
This commit is contained in:
Laisky.Cai
2024-09-25 09:22:57 +00:00
parent 4e3782d524 fdd7bf41c0
commit 3f905ea0a4
39 changed files with 873 additions and 97 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
relay/adaptor/ali/main.go
View File

@@ -9,6 +9,7 @@ import (
"github.com/gin-gonic/gin"
"github.com/songquanpeng/one-api/common"
"github.com/songquanpeng/one-api/common/ctxkey"
"github.com/songquanpeng/one-api/common/helper"
"github.com/songquanpeng/one-api/common/logger"
"github.com/songquanpeng/one-api/common/render"
@@ -59,7 +60,7 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
func ConvertEmbeddingRequest(request model.GeneralOpenAIRequest) *EmbeddingRequest {
return &EmbeddingRequest{
Model: "text-embedding-v1",
Model: request.Model,
Input: struct {
Texts []string `json:"texts"`
}{
@@ -102,8 +103,9 @@ func EmbeddingHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStat
StatusCode: resp.StatusCode,
}, nil
}
requestModel := c.GetString(ctxkey.RequestModel)
fullTextResponse := embeddingResponseAli2OpenAI(&aliResponse)
fullTextResponse.Model = requestModel
jsonResponse, err := json.Marshal(fullTextResponse)
if err != nil {
return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil

2
relay/adaptor/openai/constants.go
View File

@@ -8,6 +8,8 @@ var ModelList = []string{
"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",
"gpt-4o-2024-08-06",
"chatgpt-4o-latest",
"gpt-4o-mini", "gpt-4o-mini-2024-07-18",
"gpt-4-vision-preview",
"text-embedding-ada-002", "text-embedding-3-small", "text-embedding-3-large",

4
relay/adaptor/openai/main.go
View File

@@ -54,8 +54,8 @@ func StreamHandler(c *gin.Context, resp *http.Response, relayMode int) (*model.E
render.StringData(c, data) // if error happened, pass the data to client
continue // just ignore the error
}
if len(streamResponse.Choices) == 0 {
// but for empty choice, we should not pass it to client, this is for azure
if len(streamResponse.Choices) == 0 && streamResponse.Usage == nil {
// but for empty choice and no usage, we should not pass it to client, this is for azure
continue // just ignore empty choice
}
render.StringData(c, data)

8
relay/adaptor/stepfun/constants.go
View File

@@ -1,7 +1,13 @@
package stepfun
var ModelList = []string{
"step-1-8k",
"step-1-32k",
"step-1-128k",
"step-1-256k",
"step-1-flash",
"step-2-16k",
"step-1v-8k",
"step-1v-32k",
"step-1-200k",
"step-1x-medium",
}

1
relay/adaptor/tencent/constants.go
View File

@@ -5,4 +5,5 @@ var ModelList = []string{
"hunyuan-standard",
"hunyuan-standard-256K",
"hunyuan-pro",
"hunyuan-vision",
}

1
relay/adaptor/xunfei/constants.go
View File

@@ -5,6 +5,7 @@ var ModelList = []string{
"SparkDesk-v1.1",
"SparkDesk-v2.1",
"SparkDesk-v3.1",
"SparkDesk-v3.1-128K",
"SparkDesk-v3.5",
"SparkDesk-v4.0",
}

17
relay/adaptor/xunfei/main.go
View File

@@ -272,9 +272,9 @@ func xunfeiMakeRequest(textRequest model.GeneralOpenAIRequest, domain, authUrl,
}
func parseAPIVersionByModelName(modelName string) string {
parts := strings.Split(modelName, "-")
if len(parts) == 2 {
return parts[1]
index := strings.IndexAny(modelName, "-")
if index != -1 {
return modelName[index+1:]
}
return ""
}
@@ -288,6 +288,8 @@ func apiVersion2domain(apiVersion string) string {
return "generalv2"
case "v3.1":
return "generalv3"
case "v3.1-128K":
return "pro-128k"
case "v3.5":
return "generalv3.5"
case "v4.0":
@@ -297,7 +299,14 @@ func apiVersion2domain(apiVersion string) string {
}
func getXunfeiAuthUrl(apiVersion string, apiKey string, apiSecret string) (string, string) {
var authUrl string
domain := apiVersion2domain(apiVersion)
authUrl := buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/%s/chat", apiVersion), apiKey, apiSecret)
switch apiVersion {
case "v3.1-128K":
authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/%s/pro-128k", apiVersion), apiKey, apiSecret)
break
default:
authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/%s/chat", apiVersion), apiKey, apiSecret)
}
return domain, authUrl
}

10
relay/billing/ratio/image.go
View File

@@ -30,6 +30,14 @@ var ImageSizeRatios = map[string]map[string]float64{
"720x1280": 1,
"1280x720": 1,
},
"step-1x-medium": {
"256x256": 1,
"512x512": 1,
"768x768": 1,
"1024x1024": 1,
"1280x800": 1,
"800x1280": 1,
},
}
var ImageGenerationAmounts = map[string][2]int{
@@ -39,6 +47,7 @@ var ImageGenerationAmounts = map[string][2]int{
"ali-stable-diffusion-v1.5": {1, 4}, // Ali
"wanx-v1": {1, 4}, // Ali
"cogview-3": {1, 1},
"step-1x-medium": {1, 1},
}
var ImagePromptLengthLimitations = map[string]int{
@@ -48,6 +57,7 @@ var ImagePromptLengthLimitations = map[string]int{
"ali-stable-diffusion-v1.5": 4000,
"wanx-v1": 4000,
"cogview-3": 833,
"step-1x-medium": 4000,
}
var ImageOriginModelName = map[string]string{

27
relay/billing/ratio/model.go
View File

@@ -34,7 +34,9 @@ var ModelRatio = map[string]float64{
"gpt-4-turbo": 5, // $0.01 / 1K tokens
"gpt-4-turbo-2024-04-09": 5, // $0.01 / 1K tokens
"gpt-4o": 2.5, // $0.005 / 1K tokens
"chatgpt-4o-latest": 2.5, // $0.005 / 1K tokens
"gpt-4o-2024-05-13": 2.5, // $0.005 / 1K tokens
"gpt-4o-2024-08-06": 1.25, // $0.0025 / 1K tokens
"gpt-4o-mini": 0.075, // $0.00015 / 1K tokens
"gpt-4o-mini-2024-07-18": 0.075, // $0.00015 / 1K tokens
"gpt-4-vision-preview": 5, // $0.01 / 1K tokens
@@ -126,6 +128,7 @@ var ModelRatio = map[string]float64{
"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-v4.0": 1.2858, // ￥0.018 / 1k tokens
"360GPT_S2_V9": 0.8572, // ¥0.012 / 1k tokens
@@ -171,10 +174,15 @@ var ModelRatio = map[string]float64{
"yi-34b-chat-0205": 2.5 / 1000 * RMB,
"yi-34b-chat-200k": 12.0 / 1000 * RMB,
"yi-vl-plus": 6.0 / 1000 * RMB,
// stepfun todo
"step-1v-32k": 0.024 * RMB,
"step-1-32k": 0.024 * RMB,
"step-1-200k": 0.15 * RMB,
// https://platform.stepfun.com/docs/pricing/details
"step-1-8k": 0.005 / 1000 * RMB,
"step-1-32k": 0.015 / 1000 * RMB,
"step-1-128k": 0.040 / 1000 * RMB,
"step-1-256k": 0.095 / 1000 * RMB,
"step-1-flash": 0.001 / 1000 * RMB,
"step-2-16k": 0.038 / 1000 * RMB,
"step-1v-8k": 0.005 / 1000 * RMB,
"step-1v-32k": 0.015 / 1000 * RMB,
// aws llama3 https://aws.amazon.com/cn/bedrock/pricing/
"llama3-8b-8192(33)": 0.0003 / 0.002, // $0.0003 / 1K tokens
"llama3-70b-8192(33)": 0.00265 / 0.002, // $0.00265 / 1K tokens
@@ -200,8 +208,10 @@ var CompletionRatio = map[string]float64{
"llama3-70b-8192(33)": 0.0035 / 0.00265,
}
var DefaultModelRatio map[string]float64
var DefaultCompletionRatio map[string]float64
var (
DefaultModelRatio map[string]float64
DefaultCompletionRatio map[string]float64
)
func init() {
DefaultModelRatio = make(map[string]float64)
@@ -313,7 +323,7 @@ func GetCompletionRatio(name string, channelType int) float64 {
return 4.0 / 3.0
}
if strings.HasPrefix(name, "gpt-4") {
if strings.HasPrefix(name, "gpt-4o-mini") {
if strings.HasPrefix(name, "gpt-4o-mini") || name == "gpt-4o-2024-08-06" {
return 4
}
if strings.HasPrefix(name, "gpt-4-turbo") ||
@@ -323,6 +333,9 @@ func GetCompletionRatio(name string, channelType int) float64 {
}
return 2
}
if name == "chatgpt-4o-latest" {
return 3
}
if strings.HasPrefix(name, "claude-3") {
return 5
}

10
relay/model/general.go
View File

@@ -1,7 +1,15 @@
package model
type ResponseFormat struct {
Type string `json:"type,omitempty"`
Type string `json:"type,omitempty"`
JsonSchema *JSONSchema `json:"json_schema,omitempty"`
}
type JSONSchema struct {
Description string `json:"description,omitempty"`
Name string `json:"name"`
Schema map[string]interface{} `json:"schema,omitempty"`
Strict *bool `json:"strict,omitempty"`
}
type GeneralOpenAIRequest struct {