Merge remote-tracking branch 'origin/upstream/main'

relay/adaptor/ali/main.go

@@ -36,9 +36,7 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
 		enableSearch = true
 		aliModel = strings.TrimSuffix(aliModel, EnableSearchModelSuffix)
 	}
-	if request.TopP >= 1 {
-		request.TopP = 0.9999
-	}
+	request.TopP = helper.Float64PtrMax(request.TopP, 0.9999)
 	return &ChatRequest{
 		Model: aliModel,
 		Input: Input{

relay/adaptor/ali/model.go

@@ -16,13 +16,13 @@ type Input struct {
 }
 
 type Parameters struct {
-	TopP              float64      `json:"top_p,omitempty"`
+	TopP              *float64     `json:"top_p,omitempty"`
 	TopK              int          `json:"top_k,omitempty"`
 	Seed              uint64       `json:"seed,omitempty"`
 	EnableSearch      bool         `json:"enable_search,omitempty"`
 	IncrementalOutput bool         `json:"incremental_output,omitempty"`
 	MaxTokens         int          `json:"max_tokens,omitempty"`
-	Temperature       float64      `json:"temperature,omitempty"`
+	Temperature       *float64     `json:"temperature,omitempty"`
 	ResultFormat      string       `json:"result_format,omitempty"`
 	Tools             []model.Tool `json:"tools,omitempty"`
 }

relay/adaptor/anthropic/constants.go

@@ -3,6 +3,7 @@ package anthropic
 var ModelList = []string{
 	"claude-instant-1.2", "claude-2.0", "claude-2.1",
 	"claude-3-haiku-20240307",
+	"claude-3-5-haiku-20241022",
 	"claude-3-sonnet-20240229",
 	"claude-3-opus-20240229",
 	"claude-3-5-sonnet-20240620",

relay/adaptor/anthropic/model.go

@@ -48,8 +48,8 @@ type Request struct {
 	MaxTokens     int       `json:"max_tokens,omitempty"`
 	StopSequences []string  `json:"stop_sequences,omitempty"`
 	Stream        bool      `json:"stream,omitempty"`
-	Temperature   float64   `json:"temperature,omitempty"`
-	TopP          float64   `json:"top_p,omitempty"`
+	Temperature   *float64  `json:"temperature,omitempty"`
+	TopP          *float64  `json:"top_p,omitempty"`
 	TopK          int       `json:"top_k,omitempty"`
 	Tools         []Tool    `json:"tools,omitempty"`
 	ToolChoice    any       `json:"tool_choice,omitempty"`

relay/adaptor/aws/claude/model.go

@@ -11,8 +11,8 @@ type Request struct {
 	Messages         []anthropic.Message `json:"messages"`
 	System           string              `json:"system,omitempty"`
 	MaxTokens        int                 `json:"max_tokens,omitempty"`
-	Temperature      float64             `json:"temperature,omitempty"`
-	TopP             float64             `json:"top_p,omitempty"`
+	Temperature      *float64            `json:"temperature,omitempty"`
+	TopP             *float64            `json:"top_p,omitempty"`
 	TopK             int                 `json:"top_k,omitempty"`
 	StopSequences    []string            `json:"stop_sequences,omitempty"`
 	Tools            []anthropic.Tool    `json:"tools,omitempty"`

relay/adaptor/aws/llama3/model.go

@@ -4,10 +4,10 @@ package aws
 //
 // https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-meta.html
 type Request struct {
-	Prompt      string  `json:"prompt"`
-	MaxGenLen   int     `json:"max_gen_len,omitempty"`
-	Temperature float64 `json:"temperature,omitempty"`
-	TopP        float64 `json:"top_p,omitempty"`
+	Prompt      string   `json:"prompt"`
+	MaxGenLen   int      `json:"max_gen_len,omitempty"`
+	Temperature *float64 `json:"temperature,omitempty"`
+	TopP        *float64 `json:"top_p,omitempty"`
 }
 
 // Response is the response from AWS Llama3

relay/adaptor/baidu/main.go

@@ -35,9 +35,9 @@ type Message struct {
 
 type ChatRequest struct {
 	Messages        []Message `json:"messages"`
-	Temperature     float64   `json:"temperature,omitempty"`
-	TopP            float64   `json:"top_p,omitempty"`
-	PenaltyScore    float64   `json:"penalty_score,omitempty"`
+	Temperature     *float64  `json:"temperature,omitempty"`
+	TopP            *float64  `json:"top_p,omitempty"`
+	PenaltyScore    *float64  `json:"penalty_score,omitempty"`
 	Stream          bool      `json:"stream,omitempty"`
 	System          string    `json:"system,omitempty"`
 	DisableSearch   bool      `json:"disable_search,omitempty"`

relay/adaptor/cloudflare/model.go

@@ -9,5 +9,5 @@ type Request struct {
 	Prompt      string          `json:"prompt,omitempty"`
 	Raw         bool            `json:"raw,omitempty"`
 	Stream      bool            `json:"stream,omitempty"`
-	Temperature float64         `json:"temperature,omitempty"`
+	Temperature *float64        `json:"temperature,omitempty"`
 }

relay/adaptor/cohere/main.go

@@ -43,7 +43,7 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request {
 		K:                textRequest.TopK,
 		Stream:           textRequest.Stream,
 		FrequencyPenalty: textRequest.FrequencyPenalty,
-		PresencePenalty:  textRequest.FrequencyPenalty,
+		PresencePenalty:  textRequest.PresencePenalty,
 		Seed:             int(textRequest.Seed),
 	}
 	if cohereRequest.Model == "" {

relay/adaptor/cohere/model.go

@@ -10,15 +10,15 @@ type Request struct {
 	PromptTruncation string        `json:"prompt_truncation,omitempty"` // 默认值为"AUTO"
 	Connectors       []Connector   `json:"connectors,omitempty"`
 	Documents        []Document    `json:"documents,omitempty"`
-	Temperature      float64       `json:"temperature,omitempty"` // 默认值为0.3
+	Temperature      *float64      `json:"temperature,omitempty"` // 默认值为0.3
 	MaxTokens        int           `json:"max_tokens,omitempty"`
 	MaxInputTokens   int           `json:"max_input_tokens,omitempty"`
 	K                int           `json:"k,omitempty"` // 默认值为0
-	P                float64       `json:"p,omitempty"` // 默认值为0.75
+	P                *float64      `json:"p,omitempty"` // 默认值为0.75
 	Seed             int           `json:"seed,omitempty"`
 	StopSequences    []string      `json:"stop_sequences,omitempty"`
-	FrequencyPenalty float64       `json:"frequency_penalty,omitempty"` // 默认值为0.0
-	PresencePenalty  float64       `json:"presence_penalty,omitempty"`  // 默认值为0.0
+	FrequencyPenalty *float64      `json:"frequency_penalty,omitempty"` // 默认值为0.0
+	PresencePenalty  *float64      `json:"presence_penalty,omitempty"`  // 默认值为0.0
 	Tools            []Tool        `json:"tools,omitempty"`
 	ToolResults      []ToolResult  `json:"tool_results,omitempty"`
 }

relay/adaptor/gemini/model.go

@@ -67,8 +67,8 @@ 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"`
+	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"`

relay/adaptor/ollama/model.go

@@ -1,14 +1,14 @@
 package ollama
 
 type Options struct {
-	Seed             int     `json:"seed,omitempty"`
-	Temperature      float64 `json:"temperature,omitempty"`
-	TopK             int     `json:"top_k,omitempty"`
-	TopP             float64 `json:"top_p,omitempty"`
-	FrequencyPenalty float64 `json:"frequency_penalty,omitempty"`
-	PresencePenalty  float64 `json:"presence_penalty,omitempty"`
-	NumPredict  	 int 	 `json:"num_predict,omitempty"`
-	NumCtx  	 int 	 `json:"num_ctx,omitempty"`
+	Seed             int      `json:"seed,omitempty"`
+	Temperature      *float64 `json:"temperature,omitempty"`
+	TopK             int      `json:"top_k,omitempty"`
+	TopP             *float64 `json:"top_p,omitempty"`
+	FrequencyPenalty *float64 `json:"frequency_penalty,omitempty"`
+	PresencePenalty  *float64 `json:"presence_penalty,omitempty"`
+	NumPredict       int      `json:"num_predict,omitempty"`
+	NumCtx           int      `json:"num_ctx,omitempty"`
 }
 
 type Message struct {

relay/adaptor/openai/compatible.go

@@ -11,9 +11,10 @@ import (
 	"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/siliconflow"
 	"github.com/songquanpeng/one-api/relay/adaptor/stepfun"
 	"github.com/songquanpeng/one-api/relay/adaptor/togetherai"
-	"github.com/songquanpeng/one-api/relay/adaptor/siliconflow"
+	"github.com/songquanpeng/one-api/relay/adaptor/xai"
 	"github.com/songquanpeng/one-api/relay/channeltype"
 )
 
@@ -32,6 +33,7 @@ var CompatibleChannels = []int{
 	channeltype.TogetherAI,
 	channeltype.Novita,
 	channeltype.SiliconFlow,
+	channeltype.XAI,
 }
 
 func GetCompatibleChannelMeta(channelType int) (string, []string) {
@@ -64,6 +66,8 @@ func GetCompatibleChannelMeta(channelType int) (string, []string) {
 		return "novita", novita.ModelList
 	case channeltype.SiliconFlow:
 		return "siliconflow", siliconflow.ModelList
+	case channeltype.XAI:
+		return "xai", xai.ModelList
 	default:
 		return "openai", ModelList
 	}

relay/adaptor/palm/model.go

@@ -19,11 +19,11 @@ type Prompt struct {
 }
 
 type ChatRequest struct {
-	Prompt         Prompt  `json:"prompt"`
-	Temperature    float64 `json:"temperature,omitempty"`
-	CandidateCount int     `json:"candidateCount,omitempty"`
-	TopP           float64 `json:"topP,omitempty"`
-	TopK           int     `json:"topK,omitempty"`
+	Prompt         Prompt   `json:"prompt"`
+	Temperature    *float64 `json:"temperature,omitempty"`
+	CandidateCount int      `json:"candidateCount,omitempty"`
+	TopP           *float64 `json:"topP,omitempty"`
+	TopK           int      `json:"topK,omitempty"`
 }
 
 type Error struct {

relay/adaptor/tencent/main.go

@@ -39,8 +39,8 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
 		Model:       &request.Model,
 		Stream:      &request.Stream,
 		Messages:    messages,
-		TopP:        &request.TopP,
-		Temperature: &request.Temperature,
+		TopP:        request.TopP,
+		Temperature: request.Temperature,
 	}
 }
 

relay/adaptor/vertexai/claude/model.go

@@ -11,8 +11,8 @@ type Request struct {
 	MaxTokens     int                 `json:"max_tokens,omitempty"`
 	StopSequences []string            `json:"stop_sequences,omitempty"`
 	Stream        bool                `json:"stream,omitempty"`
-	Temperature   float64             `json:"temperature,omitempty"`
-	TopP          float64             `json:"top_p,omitempty"`
+	Temperature   *float64            `json:"temperature,omitempty"`
+	TopP          *float64            `json:"top_p,omitempty"`
 	TopK          int                 `json:"top_k,omitempty"`
 	Tools         []anthropic.Tool    `json:"tools,omitempty"`
 	ToolChoice    any                 `json:"tool_choice,omitempty"`

relay/adaptor/xai/constants.go

@@ -0,0 +1,5 @@
+package xai
+
+var ModelList = []string{
+	"grok-beta",
+}

relay/adaptor/xunfei/main.go

@@ -283,7 +283,7 @@ func parseAPIVersionByModelName(modelName string) string {
 func apiVersion2domain(apiVersion string) string {
 	switch apiVersion {
 	case "v1.1":
-		return "general"
+		return "lite"
 	case "v2.1":
 		return "generalv2"
 	case "v3.1":

relay/adaptor/xunfei/model.go

@@ -19,11 +19,11 @@ type ChatRequest struct {
 	} `json:"header"`
 	Parameter struct {
 		Chat struct {
-			Domain      string  `json:"domain,omitempty"`
-			Temperature float64 `json:"temperature,omitempty"`
-			TopK        int     `json:"top_k,omitempty"`
-			MaxTokens   int     `json:"max_tokens,omitempty"`
-			Auditing    bool    `json:"auditing,omitempty"`
+			Domain      string   `json:"domain,omitempty"`
+			Temperature *float64 `json:"temperature,omitempty"`
+			TopK        int      `json:"top_k,omitempty"`
+			MaxTokens   int      `json:"max_tokens,omitempty"`
+			Auditing    bool     `json:"auditing,omitempty"`
 		} `json:"chat"`
 	} `json:"parameter"`
 	Payload struct {

relay/adaptor/zhipu/adaptor.go

@@ -4,11 +4,11 @@ import (
 	"errors"
 	"fmt"
 	"io"
-	"math"
 	"net/http"
 	"strings"
 
 	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/relay/adaptor"
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/meta"
@@ -66,13 +66,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 		baiduEmbeddingRequest, err := ConvertEmbeddingRequest(*request)
 		return baiduEmbeddingRequest, err
 	default:
-		// TopP (0.0, 1.0)
-		request.TopP = math.Min(0.99, request.TopP)
-		request.TopP = math.Max(0.01, request.TopP)
+		// TopP [0.0, 1.0]
+		request.TopP = helper.Float64PtrMax(request.TopP, 1)
+		request.TopP = helper.Float64PtrMin(request.TopP, 0)
 
-		// Temperature (0.0, 1.0)
-		request.Temperature = math.Min(0.99, request.Temperature)
-		request.Temperature = math.Max(0.01, request.Temperature)
+		// Temperature [0.0, 1.0]
+		request.Temperature = helper.Float64PtrMax(request.Temperature, 1)
+		request.Temperature = helper.Float64PtrMin(request.Temperature, 0)
 		a.SetVersionByModeName(request.Model)
 		if a.APIVersion == "v4" {
 			return request, nil

relay/adaptor/zhipu/model.go

@@ -13,8 +13,8 @@ type Message struct {
 
 type Request struct {
 	Prompt      []Message `json:"prompt"`
-	Temperature float64   `json:"temperature,omitempty"`
-	TopP        float64   `json:"top_p,omitempty"`
+	Temperature *float64  `json:"temperature,omitempty"`
+	TopP        *float64  `json:"top_p,omitempty"`
 	RequestId   string    `json:"request_id,omitempty"`
 	Incremental bool      `json:"incremental,omitempty"`
 }