Merge branch 'dev未优化' of https://gitea.redpowerfuture.com/red-future/model-gateway into dev未优化

重新启用任务并发限制逻辑，使用 Redis 计数器控制模型并发数；
引入分布式锁防止重复创建，并优化优雅退出顺序确保请求完成；
移除 video_duration 计费计算，支持外部传入 taskId，为轮询查询添加独立超时上下文。

.dockerignore

@@ -0,0 +1 @@
+.git

Dockerfile

@@ -1,7 +1,8 @@
 # 阶段1: 构建
 FROM golang:alpine AS builder
 
-RUN apk add --no-cache git ca-certificates tzdata
+RUN sed -i 's/dl-cdn.alpinelinux.org/mirrors.aliyun.com/g' /etc/apk/repositories && \
+      apk add --no-cache git ca-certificates tzdata
 
 ENV TZ=Asia/Shanghai
 RUN ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone

common/util/billing.go

@@ -0,0 +1,332 @@
+package util
+
+import (
+	"context"
+	"fmt"
+	"math"
+	"strings"
+
+	"github.com/gogf/gf/v2/encoding/gjson"
+	"github.com/gogf/gf/v2/util/gconv"
+)
+
+// ======================== 计费入口 ========================
+
+func CalculateBilling(config map[string]any, billingData map[string]any) map[string]any {
+	if len(config) == 0 {
+		return nil
+	}
+	switch config["type"] {
+	case "inference_tier": //推理模型计费
+		return calculateInferenceTierBilling(config, billingData)
+	case "video_resolution": //视频模型计费
+		return calculateVideoResolutionBilling(config, billingData)
+	case "tts":
+		return calculateTTSBilling(config, billingData)
+	}
+	return nil
+}
+
+// ======================== 推理模型计费 ========================
+
+func calculateInferenceTierBilling(config map[string]any, data map[string]any) map[string]any {
+	promptTokens := gconv.Int64(data["prompt_tokens"])
+	completionTokens := gconv.Int64(data["completion_tokens"])
+	hasAudio := gconv.Bool(data["has_audio"])
+	inputK := promptTokens / 1000
+	tiers := config["pricing"].(map[string]any)["tiers"].([]any)
+	var matched map[string]any
+	for _, t := range tiers {
+		tier := t.(map[string]any)
+		if inputK >= gconv.Int64(tier["input_min"]) && inputK <= gconv.Int64(tier["input_max"]) {
+			matched = tier
+			break
+		}
+	}
+	if matched == nil {
+		return nil
+	}
+
+	var inputPrice float64
+	if hasAudio && matched["audio_input_price"] != nil {
+		inputPrice = gconv.Float64(matched["audio_input_price"])
+	} else {
+		inputPrice = gconv.Float64(matched["input_price"])
+	}
+	outputPrice := gconv.Float64(matched["output_price"])
+	inputCost := float64(promptTokens) * inputPrice / 1000000
+	outputCost := float64(completionTokens) * outputPrice / 1000000
+
+	// 推理模型
+	return map[string]any{
+		"model_name":   data["model_name"],
+		"total_tokens": promptTokens + completionTokens,
+		"total_fee":    inputCost + outputCost,
+		// 明细
+		"prompt_tokens":     promptTokens,
+		"completion_tokens": completionTokens,
+		"has_audio":         hasAudio,
+		"input_tier":        fmt.Sprintf("[%v, %v]", matched["input_min"], matched["input_max"]),
+		"input_unit_price":  inputPrice,
+		"output_unit_price": outputPrice,
+		"input_cost":        inputCost,
+		"output_cost":       outputCost,
+	}
+}
+
+// ======================== 视频模型计费 ========================
+
+func calculateVideoResolutionBilling(config map[string]any, data map[string]any) map[string]any {
+	pricingPath := buildPricingPath(config, data)
+
+	pricing := config["pricing"].(map[string]any)
+	unitPrice := gconv.Float64(pricing[pricingPath])
+
+	var effectiveMinToken float64
+	if gconv.Bool(data["input_has_video"]) {
+		effectiveMinToken = matchMinToken(config, data)
+	}
+
+	completionTokens := gconv.Float64(data["actual_tokens"])
+	realChargeTokens := int64(math.Max(completionTokens, effectiveMinToken))
+	totalFee := float64(realChargeTokens) * unitPrice / 1000000
+
+	if gconv.Bool(config["enable_audio_charge"]) {
+		totalFee += gconv.Float64(data["audio_word_cnt"]) * gconv.Float64(config["audio_unit_price"])
+	}
+
+	// 视频模型
+	return map[string]any{
+		"model_name":   data["model_name"],
+		"total_tokens": realChargeTokens,
+		"total_fee":    totalFee,
+		// 明细
+		"prompt_tokens":       0,
+		"completion_tokens":   int64(completionTokens),
+		"is_online":           data["is_online"],
+		"video_resolution":    data["video_resolution"],
+		"input_has_video":     data["input_has_video"],
+		"output_duration_sec": data["output_duration_sec"],
+		"aspect_ratio":        data["aspect_ratio"],
+		"resolution":          data["resolution"],
+		"matched_path":        pricingPath,
+		"token_unit_price":    unitPrice,
+		"effective_min_token": effectiveMinToken,
+	}
+}
+
+func calculateTTSBilling(config map[string]any, data map[string]any) map[string]any {
+	usage := gconv.Float64(data["synthesize_text_length"])
+	unitPrice := gconv.Float64(config["pricing"])
+	totalFee := usage * unitPrice
+
+	return map[string]any{
+		"model_name":   data["model_name"],
+		"total_tokens": int64(usage),
+		"total_fee":    totalFee,
+		// 明细
+		"prompt_tokens":     0,
+		"completion_tokens": int64(usage),
+		"unit_price":        unitPrice,
+	}
+}
+
+// ======================== 数据提取 ========================
+
+func ExtractRequestBilling(ctx context.Context, config map[string]any, requestPayload map[string]any) map[string]any {
+	dimensions := config["dimensions"].(map[string]any)
+	fields := dimensions["request"].(map[string]any)
+	data := make(map[string]any)
+
+	for key, path := range fields {
+		val := extractValue(requestPayload, gconv.String(path))
+		// TTS 模型：input_chars 自动计算字符数
+		if key == "input_chars" {
+			if s, ok := val.(string); ok {
+				data[key] = len([]rune(s))
+				continue
+			}
+		}
+		data[key] = val
+	}
+
+	if defaults, ok := config["defaults"].(map[string]any); ok {
+		for k, v := range defaults {
+			if _, exists := data[k]; !exists || data[k] == nil {
+				data[k] = v
+			}
+		}
+	}
+
+	//if compute, ok := config["compute"].(map[string]any); ok {
+	//	for targetField, rule := range compute {
+	//		r := rule.(map[string]any)
+	//		dependsOn := gconv.String(r["depends_on"])
+	//		dependsValue := gconv.Bool(r["depends_value"])
+	//
+	//		if gconv.Bool(data[dependsOn]) != dependsValue {
+	//			continue
+	//		}
+	//
+	//		switch r["service"] {
+	//		case "video_duration":
+	//			urls := extractVideoUrls(requestPayload)
+	//			if len(urls) > 0 {
+	//				resp, err := gateway.GetVideoDuration(ctx, urls)
+	//				if err == nil {
+	//					data[targetField] = resp.TotalDuration
+	//				}
+	//			}
+	//		}
+	//	}
+	//}
+	return data
+}
+
+func ExtractResponseBilling(config map[string]any, response map[string]any) map[string]any {
+	dimensions := config["dimensions"].(map[string]any)
+	fields := dimensions["response"].(map[string]any)
+	data := make(map[string]any)
+
+	for key, path := range fields {
+		data[key] = gjson.New(response).Get(gconv.String(path)).Val()
+	}
+	return data
+}
+
+// ======================== 内部辅助 ========================
+
+func buildPricingPath(config map[string]any, data map[string]any) string {
+	pathConfig := config["pricing_path"].(map[string]any)
+	segments := pathConfig["segments"].([]any)
+	mapping := pathConfig["mapping"].(map[string]any)
+
+	var parts []string
+	for _, seg := range segments {
+		segName := gconv.String(seg)
+		val := gconv.String(data[segName])
+		if m, ok := mapping[segName].(map[string]any); ok {
+			if mapped, exists := m[val]; exists {
+				val = gconv.String(mapped)
+			}
+		}
+		parts = append(parts, val)
+	}
+	return strings.Join(parts, ".")
+}
+
+func matchMinToken(config map[string]any, data map[string]any) float64 {
+	rule, ok := config["min_token_rule"].(map[string]any)
+	if !ok {
+		return 0
+	}
+
+	conditionDefs := rule["conditions"].([]any)
+	rows := rule["rows"].([]any)
+
+	for _, row := range rows {
+		r := row.(map[string]any)
+		conditions := r["conditions"].([]any)
+		matched := true
+
+		for i, cond := range conditions {
+			condStr := gconv.String(cond)
+			condDef := conditionDefs[i].(map[string]any)
+			condName := gconv.String(condDef["name"])
+
+			switch condDef["type"] {
+			case "range":
+				if !matchRange(condStr, gconv.Float64(data[condName])) {
+					matched = false
+				}
+			case "enum":
+				if gconv.String(data[condName]) != condStr {
+					matched = false
+				}
+			}
+		}
+		if matched {
+			return gconv.Float64(r["min_token"])
+		}
+	}
+	return 0
+}
+
+func matchRange(rangeStr string, val float64) bool {
+	rangeStr = strings.Trim(rangeStr, "[]()")
+	parts := strings.Split(rangeStr, ",")
+	if len(parts) != 2 {
+		return false
+	}
+	return val >= gconv.Float64(strings.TrimSpace(parts[0])) && val < gconv.Float64(strings.TrimSpace(parts[1]))
+}
+
+func extractValue(source map[string]any, path string) any {
+	// 数组求和：rounds.#.duration
+	if strings.HasPrefix(path, "rounds.#.") && !strings.Contains(path, "==") {
+		field := strings.TrimPrefix(path, "rounds.#.")
+		rounds := gjson.New(source).Get("rounds").Array()
+		var sum float64
+		for _, r := range rounds {
+			if strings.Contains(field, "#") {
+				parts := strings.SplitN(field, ".#.", 2)
+				arr := gjson.New(r).Get(parts[0]).Array()
+				for _, item := range arr {
+					sum += gconv.Float64(gjson.New(item).Get(parts[1]).Val())
+				}
+			} else {
+				sum += gconv.Float64(gjson.New(r).Get(field).Val())
+			}
+		}
+		return sum
+	}
+
+	// 条件判断：xxx.#.type==yyy 或 rounds.#.content.#.type==yyy
+	if strings.Contains(path, "==video_url") || strings.Contains(path, "==input_audio") {
+		parts := strings.Split(path, "==")
+		basePath := parts[0]
+		typ := parts[1]
+		segments := strings.Split(basePath, ".#.")
+
+		// 单层：content.#.type → segments = ["content", "type"]
+		if len(segments) == 2 && !strings.Contains(segments[0], ".") {
+			arr := gjson.New(source).Get(segments[0]).Array()
+			for _, item := range arr {
+				if gconv.String(gjson.New(item).Get(segments[1]).Val()) == typ {
+					return true
+				}
+			}
+			return false
+		}
+
+		// 嵌套：rounds.#.content.#.type → segments = ["rounds", "content", "type"]
+		topArr := gjson.New(source).Get(segments[0]).Array()
+		for _, item := range topArr {
+			subArr := gjson.New(item).Get(segments[1]).Array()
+			for _, sub := range subArr {
+				if gconv.String(gjson.New(sub).Get("type").Val()) == typ {
+					return true
+				}
+			}
+		}
+		return false
+	}
+
+	return gjson.New(source).Get(path).Val()
+}
+
+func extractVideoUrls(body map[string]any) []string {
+	var urls []string
+	content := gjson.New(body).Get("content").Array()
+	for _, c := range content {
+		item := c.(map[string]any)
+		if item["type"] == "video_url" {
+			if v, ok := item["video_url"].(map[string]any); ok {
+				if url, ok := v["url"].(string); ok {
+					urls = append(urls, url)
+				}
+			}
+		}
+	}
+	return urls
+}

common/util/files.go

@@ -1,7 +1,6 @@
 package util
 
 import (
-	"encoding/json"
 	"fmt"
 	"net/http"
 	"os"
@@ -9,107 +8,75 @@ import (
 	"strings"
 )
 
-// DetectFileType 根据返回的二进制内容推断 contentType + 扩展名（尽量稳定）
+// DetectFileType 根据返回的二进制内容推断 contentType + 扩展名
 func DetectFileType(data []byte) (contentType string, ext string) {
 	if len(data) == 0 {
-		return "application/octet-stream", ""
+		return "application/octet-stream", ".bin"
 	}
+
 	ct := http.DetectContentType(data)
-	// gateway.DetectContentType 可能带 charset 等参数：text/plain; charset=utf-8
 	if idx := strings.Index(ct, ";"); idx > 0 {
 		ct = strings.TrimSpace(ct[:idx])
 	}
+
 	switch ct {
 	case "audio/mpeg":
 		return ct, ".mp3"
 	case "audio/wave", "audio/wav", "audio/x-wav":
 		return ct, ".wav"
+	case "audio/mp4", "audio/x-m4a":
+		return ct, ".m4a"
 	case "video/mp4":
 		return ct, ".mp4"
+	case "video/webm":
+		return ct, ".webm"
 	case "image/png":
 		return ct, ".png"
 	case "image/jpeg":
 		return ct, ".jpg"
+	case "image/gif":
+		return ct, ".gif"
+	case "image/webp":
+		return ct, ".webp"
 	case "application/pdf":
 		return ct, ".pdf"
 	case "text/plain":
 		return ct, ".txt"
 	case "application/json":
 		return ct, ".json"
+	case "application/zip":
+		return ct, ".zip"
+	case "application/octet-stream":
+		return ct, ".bin"
 	default:
-		// 兜底：尝试从 ct 截取 subtype 作为后缀（例如 application/json）
 		if parts := strings.Split(ct, "/"); len(parts) == 2 {
 			sub := parts[1]
-			// 避免出现 "plain; charset=utf-8" 之类的后缀
 			if idx := strings.Index(sub, ";"); idx > 0 {
 				sub = strings.TrimSpace(sub[:idx])
 			}
 			return ct, "." + sub
 		}
-		return ct, ""
+		return ct, ".bin"
 	}
 }
 
-// SaveTmpResult 将模型输出写入临时文件，用于 OSS 上传失败后的“仅重试 OSS”。
+// SaveTmpResult 将二进制数据写入临时文件
 func SaveTmpResult(taskID string, data []byte, ext string) (string, error) {
 	dir := filepath.Join(os.TempDir(), "model-asynch")
 	if err := os.MkdirAll(dir, 0o755); err != nil {
-		return "", err
+		return "", fmt.Errorf("创建临时目录失败: %w", err)
 	}
+
 	if ext == "" {
 		ext = ".bin"
 	}
 	if ext[0] != '.' {
 		ext = "." + ext
 	}
+
 	path := filepath.Join(dir, fmt.Sprintf("%s%s", taskID, ext))
 	if err := os.WriteFile(path, data, 0o644); err != nil {
-		return "", err
+		return "", fmt.Errorf("写入临时文件失败: %w", err)
 	}
 	return path, nil
 }
-
-// SaveTempFileByType
-// 根据传入的数据自动判断：
-// 若是 []byte 且后缀为 .mp3 → 保存二进制音频
-// 若是任意结构体/map → 自动转 JSON 保存
-// 返回：新临时文件路径、错误
-func SaveTempFileByType(taskID string, data any, oldTmpFile string) (string, error) {
-	// 1. 先清理旧临时文件（统一逻辑）
-	if oldTmpFile != "" {
-		_ = os.Remove(oldTmpFile)
-	}
-
-	var tmpPath string
-	var tmpErr error
-
-	// 2. 判断是否是二进制音频（[]byte + .mp3）
-	if audioData, ok := data.([]byte); ok {
-		tmpPath, tmpErr = saveTmpResult(taskID, audioData, ".mp3")
-	} else {
-		// 3. 其他类型 → 序列化为 JSON 保存
-		mappedBytes, err := json.Marshal(data)
-		if err != nil {
-			return "", err
-		}
-		if len(mappedBytes) == 0 {
-			return "", nil
-		}
-		tmpPath, tmpErr = saveTmpResult(taskID, mappedBytes, ".json")
-	}
-
-	if tmpErr != nil || tmpPath == "" {
-		return "", tmpErr
-	}
-
-	return tmpPath, nil
-}
-
-// saveTmpResult 你原有的底层保存文件方法（保留不动）
-func saveTmpResult(taskID string, data []byte, ext string) (string, error) {
-	// 你原来实现，比如：
-	filename := taskID + ext
-	tmpPath := filepath.Join(os.TempDir(), filename)
-	err := os.WriteFile(tmpPath, data, 0644)
-	return tmpPath, err
-}

common/util/mapping.go

@@ -19,41 +19,59 @@ import (
 	tgjson "github.com/tidwall/gjson"
 )
 
-// ParseAndValidate 解析并校验结果
-func ParseAndValidate(raw map[string]any, model *entity.ModelGatewayModel) (map[string]any, error) {
-	// 1) 解析 content 字符串为 rounds 数组
-	contentVal, ok := raw[model.ResponseBody]
-	if !ok {
-		return raw, fmt.Errorf("字段 %s 不存在", model.ResponseBody)
-	}
-	contentStr, ok := contentVal.(string)
-	if !ok || strings.TrimSpace(contentStr) == "" {
-		return raw, fmt.Errorf("字段 %s 为空或不是字符串", model.ResponseBody)
-	}
-	var arr []any
-	if err := json.Unmarshal([]byte(contentStr), &arr); err != nil {
-		return raw, fmt.Errorf("JSON解析失败: %w", err)
-	}
-	if len(arr) == 0 {
-		return raw, fmt.Errorf("解析后数组为空")
+// ParseAndValidate 解析模型响应，并返回标准格式
+func ParseAndValidate(raw map[string]any, requiredFields []string) (map[string]any, error) {
+	contentStr := gconv.String(raw[entity.ResponseBody])
+	if strings.TrimSpace(contentStr) == "" {
+		return raw, fmt.Errorf("字段 %s 为空", entity.ResponseBody)
 	}
 
-	// 2) 校验必填字段
-	if len(model.RequiredFields) > 0 {
-		for i, r := range arr {
-			round, ok := r.(map[string]any)
+	contentStr = strings.Map(func(r rune) rune {
+		if r < 32 && r != ' ' {
+			return -1
+		}
+		return r
+	}, contentStr)
+
+	// 第一步：先解析为通用 interface{}，判断是对象还是数组
+	var data any
+	if err := json.Unmarshal([]byte(contentStr), &data); err != nil {
+		return raw, fmt.Errorf("JSON解析失败: %w", err)
+	}
+
+	var arr []any
+	switch val := data.(type) {
+	case []any:
+		// 本身就是数组，直接赋值
+		arr = val
+	case map[string]any:
+		// 单个对象，包装成单元素数组，统一后续逻辑
+		arr = []any{val}
+	default:
+		return raw, fmt.Errorf("不支持的JSON类型，仅允许对象/数组")
+	}
+
+	if len(arr) == 0 {
+		return raw, fmt.Errorf("解析后数据数组为空")
+	}
+
+	// 校验每一项的必填字段
+	for _, field := range requiredFields {
+		for i, item := range arr {
+			itemMap, ok := item.(map[string]any)
 			if !ok {
-				continue
+				return raw, fmt.Errorf("rounds[%d] 不是合法JSON对象", i)
 			}
-			for _, field := range model.RequiredFields {
-				if gjson.New(round).Get(field).IsNil() {
-					return raw, fmt.Errorf("rounds[%d] 缺少必填字段: %s", i, field)
-				}
+			if gjson.New(itemMap).Get(field).IsNil() {
+				return raw, fmt.Errorf("rounds[%d] 缺少必填字段: %s", i, field)
 			}
 		}
 	}
 
-	return map[string]any{"total_rounds": len(arr), "rounds": arr}, nil
+	return map[string]any{
+		"total_rounds": len(arr),
+		"rounds":       arr,
+	}, nil
 }
 
 // ParseStructResult 解析结构结果
@@ -266,10 +284,17 @@ func PullTaskResult(ctx context.Context, body map[string]any, queryConfig map[st
 		}
 
 		if matchStatus(statusStr, statusValues["failed"]) {
-			g.Log().Errorf(ctx, "[PullTaskResult] 任务失败 taskID=%s", taskID)
-			return result, fmt.Errorf("任务失败")
+			errMsg := gconv.String(gjson.New(result).Get("error.message").Val())
+			if errMsg == "" {
+				errMsg = gconv.String(gjson.New(result).Get("error").Val())
+			}
+			if errMsg == "" {
+				rawBytes, _ := json.Marshal(result)
+				errMsg = string(rawBytes)
+			}
+			g.Log().Errorf(ctx, "[PullTaskResult] 任务失败 taskID=%s err=%s", taskID, errMsg)
+			return result, fmt.Errorf("任务失败: %s", errMsg)
 		}
-
 		time.Sleep(time.Duration(interval) * time.Second)
 	}
 }

config.yml

@@ -7,7 +7,7 @@ server:
 database:
   default:
     - type: "pgsql"
-      host: "116.204.74.41"
+      host: "192.168.0.83"
       port: "15432"
       user: "postgres"
       pass: "Bjang09@686^*^"
@@ -28,7 +28,7 @@ database:
       timeMaintainDisabled: false # (可选)是否完全关闭时间更新特性，为true时CreatedAt/UpdatedAt/DeletedAt都将失效
   model_gateway:
     - type: "pgsql"
-      host: "116.204.74.41"
+      host: "192.168.0.83"
       port: "15432"
       user: "postgres"
       pass: "Bjang09@686^*^"
@@ -39,8 +39,8 @@ database:
       dryRun: false
       charset: "utf8"
       timezone: "Asia/Shanghai"
-      maxIdle: 5
-      maxOpen: 20
+      maxIdle: 15
+      maxOpen: 60
       maxLifetime: "30s"
       maxIdleConnTime: "30s"
       createdAt: "created_at"
@@ -50,14 +50,14 @@ database:
 
 redis:
   default:
-    address: 192.168.3.30:6379
+    address: 192.168.0.83:6379
     db: 0
 
 consul:
-  address: 192.168.3.30:8500
+  address: 192.168.0.83:8500
 
 jaeger:
-  addr: 192.168.3.30:4318
+  addr: 192.168.0.83:4318
 
 # 本地调试用：可选自动执行 worker/cleaner（默认关闭）
 asynch:

consts/model/model_supplier.go

@@ -0,0 +1,108 @@
+package model
+
+import (
+	"model-gateway/consts/public"
+
+	"github.com/gogf/gf/v2/util/gconv"
+)
+
+// 供应商编码常量
+const (
+	SupplierAliyun     = 1
+	SupplierVolcengine = 2
+	SupplierTencent    = 3
+	SupplierHuawei     = 4
+	SupplierBaidu      = 5
+	SupplierOpenAI     = 6
+	SupplierAzure      = 7
+	SupplierAWS        = 8
+	SupplierGoogle     = 9
+	SupplierDeepSeek   = 10
+	SupplierMoonshot   = 11
+	SupplierZhipu      = 12
+	SupplierBaichuan   = 13
+	SupplierMinimax    = 14
+	SupplierXunfei     = 15
+	SupplierOthers     = 16
+)
+
+// SupplierType 供应商编码类型
+type SupplierType *int8
+
+// SupplierItem 供应商项
+type SupplierItem struct {
+	Code SupplierType `json:"code"`
+	Desc string       `json:"desc"`
+}
+
+// 名称映射【唯一文案维护】
+var supplierNameMap = map[int]string{
+	SupplierAliyun:     "阿里云百炼",
+	SupplierVolcengine: "火山引擎",
+	SupplierTencent:    "腾讯云",
+	SupplierHuawei:     "华为云",
+	SupplierBaidu:      "百度智能云",
+	SupplierOpenAI:     "OpenAI",
+	SupplierAzure:      "Azure OpenAI",
+	SupplierAWS:        "AWS Bedrock",
+	SupplierGoogle:     "Google Cloud",
+	SupplierDeepSeek:   "DeepSeek",
+	SupplierMoonshot:   "Moonshot",
+	SupplierZhipu:      "智谱AI",
+	SupplierBaichuan:   "百川智能",
+	SupplierMinimax:    "MiniMax",
+	SupplierXunfei:     "科大讯飞",
+	SupplierOthers:     "其他",
+}
+
+// 供应商展示顺序
+var supplierOrder = []int{
+	SupplierAliyun, SupplierVolcengine, SupplierTencent, SupplierHuawei, SupplierBaidu,
+	SupplierOpenAI, SupplierAzure, SupplierAWS, SupplierGoogle, SupplierDeepSeek,
+	SupplierMoonshot, SupplierZhipu, SupplierBaichuan, SupplierMinimax, SupplierXunfei, SupplierOthers,
+}
+
+// 全局供应商实例
+var (
+	SupplierItemAliyun     = newSupplierItem(gconv.PtrInt8(SupplierAliyun))
+	SupplierItemVolcengine = newSupplierItem(gconv.PtrInt8(SupplierVolcengine))
+	SupplierItemTencent    = newSupplierItem(gconv.PtrInt8(SupplierTencent))
+	SupplierItemHuawei     = newSupplierItem(gconv.PtrInt8(SupplierHuawei))
+	SupplierItemBaidu      = newSupplierItem(gconv.PtrInt8(SupplierBaidu))
+	SupplierItemOpenAI     = newSupplierItem(gconv.PtrInt8(SupplierOpenAI))
+	SupplierItemAzure      = newSupplierItem(gconv.PtrInt8(SupplierAzure))
+	SupplierItemAWS        = newSupplierItem(gconv.PtrInt8(SupplierAWS))
+	SupplierItemGoogle     = newSupplierItem(gconv.PtrInt8(SupplierGoogle))
+	SupplierItemDeepSeek   = newSupplierItem(gconv.PtrInt8(SupplierDeepSeek))
+	SupplierItemMoonshot   = newSupplierItem(gconv.PtrInt8(SupplierMoonshot))
+	SupplierItemZhipu      = newSupplierItem(gconv.PtrInt8(SupplierZhipu))
+	SupplierItemBaichuan   = newSupplierItem(gconv.PtrInt8(SupplierBaichuan))
+	SupplierItemMinimax    = newSupplierItem(gconv.PtrInt8(SupplierMinimax))
+	SupplierItemXunfei     = newSupplierItem(gconv.PtrInt8(SupplierXunfei))
+	SupplierItemOthers     = newSupplierItem(gconv.PtrInt8(SupplierOthers))
+)
+
+func newSupplierItem(code SupplierType) SupplierItem {
+	val := int(*code)
+	return SupplierItem{
+		Code: code,
+		Desc: supplierNameMap[val],
+	}
+}
+
+// GetSupplierDescByCode 根据编码获取供应商名称
+func GetSupplierDescByCode(code int) string {
+	return supplierNameMap[code]
+}
+
+// GetSupplierOptionList 获取供应商下拉列表
+func GetSupplierOptionList() []*public.Option {
+	var list []*public.Option
+	for _, code := range supplierOrder {
+		list = append(list, &public.Option{
+			Value: code,
+			Label: supplierNameMap[code],
+		})
+	}
+	return list
+}

consts/model/model_type.go

@@ -0,0 +1,186 @@
+package model
+
+import (
+	"model-gateway/consts/public"
+
+	"github.com/gogf/gf/v2/util/gconv"
+)
+
+// 模型类型编码常量
+const (
+	TypeInference = 100 // 推理模型
+	TypeImage     = 200 // 图片模型
+	TypeAudio     = 300 // 音频模型
+	TypeVector    = 400 // 向量化模型
+	TypeOmni      = 500 // 全模态模型
+	TypeVideo     = 600 // 视频模型
+
+	// 图片子类型
+	ImageSubTextToImage      = 201
+	ImageSubImageToImage     = 202
+	ImageSubImageEdit        = 203
+	ImageSubImageVariation   = 204
+	ImageSubImageTextToImage = 205
+
+	// 音频子类型
+	AudioSubTextToSpeech   = 301
+	AudioSubSpeechToText   = 302
+	AudioSubSpeechToSpeech = 303
+
+	// 向量化子类型
+	VectorSubEmbedding = 401
+	VectorSubRerank    = 402
+
+	// 全模态子类型
+	OmniSubTextImageAudio = 501
+	OmniSubVision         = 502
+
+	// 视频子类型
+	VideoSubTextToVideo      = 601
+	VideoSubImageToVideo     = 602
+	VideoSubImageTextToVideo = 603
+	VideoSubVideoToVideo     = 604
+)
+
+// ModelType 编码类型
+type ModelType *int
+
+// ModelTypeItem 模型类型项
+type ModelTypeItem struct {
+	Code ModelType `json:"code"`
+	Desc string    `json:"desc"`
+}
+
+// TypeTree 树形结构
+type TypeTree struct {
+	Value    int              `json:"value"`
+	Label    string           `json:"label"`
+	Children []*public.Option `json:"children"`
+}
+
+// 名称映射表【唯一文案维护入口】
+var typeNameMap = map[int]string{
+	TypeInference: "推理模型",
+	TypeImage:     "图片模型",
+	TypeAudio:     "音频模型",
+	TypeVector:    "向量化模型",
+	TypeOmni:      "全模态模型",
+	TypeVideo:     "视频模型",
+
+	ImageSubTextToImage:      "文生图",
+	ImageSubImageToImage:     "图生图",
+	ImageSubImageEdit:        "图片编辑",
+	ImageSubImageVariation:   "图片变体",
+	ImageSubImageTextToImage: "图文生图",
+
+	AudioSubTextToSpeech:   "文生音",
+	AudioSubSpeechToText:   "音生文",
+	AudioSubSpeechToSpeech: "音生音",
+
+	VectorSubEmbedding: "文本嵌入",
+	VectorSubRerank:    "重排序",
+
+	OmniSubTextImageAudio: "文图音",
+	OmniSubVision:         "视觉理解",
+
+	VideoSubTextToVideo:      "文生视频",
+	VideoSubImageToVideo:     "图生视频",
+	VideoSubImageTextToVideo: "图文生视频",
+	VideoSubVideoToVideo:     "视频生视频",
+}
+
+// 父子级映射（仅存有子项的分类）
+var parentChildMap = map[int][]int{
+	TypeImage:  {ImageSubTextToImage, ImageSubImageToImage, ImageSubImageEdit, ImageSubImageVariation, ImageSubImageTextToImage},
+	TypeAudio:  {AudioSubTextToSpeech, AudioSubSpeechToText, AudioSubSpeechToSpeech},
+	TypeVector: {VectorSubEmbedding, VectorSubRerank},
+	TypeOmni:   {OmniSubTextImageAudio, OmniSubVision},
+	TypeVideo:  {VideoSubTextToVideo, VideoSubImageToVideo, VideoSubImageTextToVideo, VideoSubVideoToVideo},
+}
+
+// 一级分类展示顺序
+var parentTypeOrder = []int{
+	TypeInference, TypeImage, TypeAudio, TypeVector, TypeOmni, TypeVideo,
+}
+
+// 全局实例：一级 + 全部二级子类型，统一通过 newItem 构造，文案仅维护在 typeNameMap
+var (
+	// 一级类型
+	ModelTypeInference = newItem(gconv.PtrInt(TypeInference))
+	ModelTypeImage     = newItem(gconv.PtrInt(TypeImage))
+	ModelTypeAudio     = newItem(gconv.PtrInt(TypeAudio))
+	ModelTypeVector    = newItem(gconv.PtrInt(TypeVector))
+	ModelTypeOmni      = newItem(gconv.PtrInt(TypeOmni))
+	ModelTypeVideo     = newItem(gconv.PtrInt(TypeVideo))
+
+	// 图片二级子类型
+	ModelImageSubTextToImage      = newItem(gconv.PtrInt(ImageSubTextToImage))
+	ModelImageSubImageToImage     = newItem(gconv.PtrInt(ImageSubImageToImage))
+	ModelImageSubImageEdit        = newItem(gconv.PtrInt(ImageSubImageEdit))
+	ModelImageSubImageVariation   = newItem(gconv.PtrInt(ImageSubImageVariation))
+	ModelImageSubImageTextToImage = newItem(gconv.PtrInt(ImageSubImageTextToImage))
+
+	// 音频二级子类型
+	ModelAudioSubTextToSpeech   = newItem(gconv.PtrInt(AudioSubTextToSpeech))
+	ModelAudioSubSpeechToText   = newItem(gconv.PtrInt(AudioSubSpeechToText))
+	ModelAudioSubSpeechToSpeech = newItem(gconv.PtrInt(AudioSubSpeechToSpeech))
+
+	// 向量化二级子类型
+	ModelVectorSubEmbedding = newItem(gconv.PtrInt(VectorSubEmbedding))
+	ModelVectorSubRerank    = newItem(gconv.PtrInt(VectorSubRerank))
+
+	// 全模态二级子类型
+	ModelOmniSubTextImageAudio = newItem(gconv.PtrInt(OmniSubTextImageAudio))
+	ModelOmniSubVision         = newItem(gconv.PtrInt(OmniSubVision))
+
+	// 视频二级子类型
+	ModelVideoSubTextToVideo      = newItem(gconv.PtrInt(VideoSubTextToVideo))
+	ModelVideoSubImageToVideo     = newItem(gconv.PtrInt(VideoSubImageToVideo))
+	ModelVideoSubImageTextToVideo = newItem(gconv.PtrInt(VideoSubImageTextToVideo))
+	ModelVideoSubVideoToVideo     = newItem(gconv.PtrInt(VideoSubVideoToVideo))
+)
+
+// newItem 构造方法：自动从 typeNameMap 读取描述
+func newItem(code ModelType) ModelTypeItem {
+	val := int(*code)
+	return ModelTypeItem{
+		Code: code,
+		Desc: typeNameMap[val],
+	}
+}
+
+// GetDescByCode 根据编码获取名称
+func GetDescByCode(code int) string {
+	return typeNameMap[code]
+}
+
+// GetTypeTreeList 生成树形数据
+func GetTypeTreeList() []*TypeTree {
+	var list []*TypeTree
+	for _, parentCode := range parentTypeOrder {
+		tree := &TypeTree{
+			Value:    parentCode,
+			Label:    typeNameMap[parentCode],
+			Children: make([]*public.Option, 0),
+		}
+		if childCodes, ok := parentChildMap[parentCode]; ok {
+			for _, c := range childCodes {
+				tree.Children = append(tree.Children, &public.Option{
+					Value: c,
+					Label: typeNameMap[c],
+				})
+			}
+		}
+		list = append(list, tree)
+	}
+	return list
+}
+
+// GetAllTypeOption 全量平铺选项
+func GetAllTypeOption() []*public.Option {
+	var list []*public.Option
+	for code, label := range typeNameMap {
+		list = append(list, &public.Option{Value: code, Label: label})
+	}
+	return list
+}

consts/model/response_type.go

@@ -0,0 +1,27 @@
+package model
+
+import "github.com/gogf/gf/v2/util/gconv"
+
+var (
+	ResponseTypeSync   = newResponseType(gconv.PtrInt8(1), "sync")   // 同步
+	ResponseTypeAsync  = newResponseType(gconv.PtrInt8(2), "async")  // 异步
+	ResponseTypeStream = newResponseType(gconv.PtrInt8(3), "stream") // 流
+)
+
+type ResponseType *int8
+
+type responseType struct {
+	code ResponseType
+	desc string
+}
+
+func (s responseType) Code() ResponseType {
+	return s.code
+}
+func (s responseType) Desc() string {
+	return s.desc
+}
+
+func newResponseType(code ResponseType, desc string) responseType {
+	return responseType{code: code, desc: desc}
+}

consts/public/public.go

@@ -1,5 +1,11 @@
 package public
 
+// Option 通用下拉选项
+type Option struct {
+	Value int    `json:"value"`
+	Label string `json:"label"`
+}
+
 const (
 	CallModeSync   = 0 // 同步调用
 	CallModeAsync  = 1 // 异步调用

consts/public/table_name.go

@@ -5,8 +5,9 @@ const (
 )
 
 const (
-	TableNameModel = "model_gateway_models"    // 模型表
-	TableNameTask  = "model_gateway_task"      // 任务表
-	TableNameOpLog = "model_gateway_logs_op"   // 操作日志表
-	TableNameStat  = "model_gateway_logs_stat" // 按天统计表
+	TableNameModel       = "model_gateway_models"    // 模型表
+	TableNameTask        = "model_gateway_task"      // 任务表
+	TableNameOpLog       = "model_gateway_logs_op"   // 操作日志表
+	TableNameStat        = "model_gateway_logs_stat" // 按天统计表
+	TableNameModelManage = "model_gateway_model_manage"
 )

controller/model_manage_controller.go

@@ -0,0 +1,56 @@
+package controller
+
+import (
+	"context"
+	"model-gateway/model/dto"
+	"model-gateway/service"
+
+	"gitea.redpowerfuture.com/red-future/common/beans"
+)
+
+// ModelManage 模型配置控制器
+var ModelManage = new(modelManage)
+
+type modelManage struct{}
+
+// CreateModel 添加配置
+func (c *modelManage) CreateModel(ctx context.Context, req *dto.CreateModelManageReq) (res *dto.CreateModelManageRes, err error) {
+	return service.ModelManage.Create(ctx, req)
+}
+
+// UpdateModel 更改配置
+func (c *modelManage) UpdateModel(ctx context.Context, req *dto.UpdateModelManageReq) (res *beans.ResponseEmpty, err error) {
+	err = service.ModelManage.Update(ctx, req)
+	return
+}
+
+// DeleteModel 删除配置
+func (c *modelManage) DeleteModel(ctx context.Context, req *dto.DeleteModelManageReq) (res *beans.ResponseEmpty, err error) {
+	err = service.ModelManage.Delete(ctx, req)
+	return
+}
+
+// GetModel 获取配置
+func (c *modelManage) GetModel(ctx context.Context, req *dto.GetModelManageReq) (res *dto.GetModelManageRes, err error) {
+	return service.ModelManage.Get(ctx, req)
+}
+
+// ListModel 配置列表
+func (c *modelManage) ListModel(ctx context.Context, req *dto.ListModelManageReq) (res *dto.ListModelManageRes, err error) {
+	return service.ModelManage.List(ctx, req)
+}
+
+// CheckChatModel 检查是否为聊天模型
+func (c *modelManage) CheckChatModel(ctx context.Context, req *dto.CheckChatModelReq) (res *dto.CheckChatModelRes, err error) {
+	return service.ModelManage.CheckChatModel(ctx, req)
+}
+
+// ListType 模型类型列表
+func (c *modelManage) ListType(ctx context.Context, req *dto.ModelTypeReq) (res *dto.ModelTypeRes, err error) {
+	return service.ModelManage.GetModelType(ctx, req)
+}
+
+// ListOperator 运营商列表
+func (c *modelManage) ListOperator(ctx context.Context, req *dto.ModelSupplierReq) (res *dto.ModelSupplierRes, err error) {
+	return service.ModelManage.GetModelSupplier(ctx, req)
+}

dao/model_gateway_models_dao.go

@@ -56,6 +56,7 @@ func (d *modelGatewayModelsDao) Get(ctx context.Context, req *entity.ModelGatewa
 		Where(entity.ModelGatewayModelCol.Id, req.Id).
 		Where(entity.ModelGatewayModelCol.Creator, req.Creator).
 		Where(entity.ModelGatewayModelCol.ModelName, req.ModelName).
+		Where(entity.ModelGatewayModelCol.IsChatModel, req.IsChatModel).
 		Fields(fields).One()
 	if err != nil {
 		return nil, err
@@ -122,7 +123,7 @@ func (d *modelGatewayModelsDao) GetByAcrossTenant(ctx context.Context, req *enti
 func (d *modelGatewayModelsDao) GetByCreatorAndPlatform(ctx context.Context, req *dto.ListModelReq) (list []*entity.ModelGatewayModel, total int, err error) {
 	sql := `
 SELECT DISTINCT ON (model_name) *
-FROM asynch_models
+FROM ` + public.TableNameModel + `
 WHERE deleted_at IS NULL 
   AND (? = '' OR model_name LIKE ?)
 `

dao/model_gateway_task_dao.go

@@ -7,7 +7,6 @@ import (
 	"model-gateway/model/entity"
 
 	"gitea.redpowerfuture.com/red-future/common/db/gfdb"
-	"github.com/gogf/gf/v2/database/gdb"
 	"github.com/gogf/gf/v2/util/gconv"
 )
 
@@ -128,32 +127,32 @@ func (d *modelGatewayTaskDao) GetPendingAsyncTasks(ctx context.Context, limit in
 
 // ClaimByID 按主键抢占，返回抢占后的任务
 func (d *modelGatewayTaskDao) ClaimByID(ctx context.Context, id int64) (*entity.ModelGatewayTask, error) {
+	// 1) 先查任务
 	var task entity.ModelGatewayTask
-	err := gfdb.DB(ctx, public.DbNameModelGateway).Transaction(ctx, func(ctx context.Context, tx gdb.TX) error {
-		r, err := tx.Model(public.TableNameTask).
-			Where(entity.ModelGatewayTaskCol.Id, id).
-			Where(entity.ModelGatewayTaskCol.State, public.TaskStatusPending).
-			Limit(1).
-			LockUpdate().
-			One()
-		if err != nil {
-			return err
-		}
-		if r.IsEmpty() {
-			return fmt.Errorf("任务已被抢占或不存在: id=%d", id)
-		}
-		if err := r.Struct(&task); err != nil {
-			return err
-		}
-		_, err = tx.Model(public.TableNameTask).
-			Data(&entity.ModelGatewayTask{State: public.TaskStatusRunning}).
-			Where(entity.ModelGatewayTaskCol.Id, id).
-			OmitEmpty().
-			Update()
-		return err
-	})
+	r, err := gfdb.DB(ctx, public.DbNameModelGateway).Model(ctx, public.TableNameTask).
+		Where(entity.ModelGatewayTaskCol.Id, id).
+		Where(entity.ModelGatewayTaskCol.State, public.TaskStatusPending).
+		One()
 	if err != nil {
 		return nil, err
 	}
+	if r.IsEmpty() {
+		return nil, fmt.Errorf("任务已被抢占或不存在: id=%d", id)
+	}
+	if err = r.Struct(&task); err != nil {
+		return nil, err
+	}
+
+	// 2) 改为执行中
+	_, err = gfdb.DB(ctx, public.DbNameModelGateway).Model(ctx, public.TableNameTask).
+		Data(&entity.ModelGatewayTask{State: public.TaskStatusRunning}).
+		Where(entity.ModelGatewayTaskCol.Id, id).
+		Where(entity.ModelGatewayTaskCol.State, public.TaskStatusPending). // 防并发
+		OmitEmpty().
+		Update()
+	if err != nil {
+		return nil, err
+	}
+
 	return &task, nil
 }

dao/model_manage_dao.go

@@ -0,0 +1,193 @@
+package dao
+
+import (
+	"context"
+	"fmt"
+	"model-gateway/consts/public"
+	"model-gateway/model/dto"
+	"model-gateway/model/entity"
+
+	"gitea.redpowerfuture.com/red-future/common/db/gfdb"
+	"github.com/gogf/gf/v2/frame/g"
+	"github.com/gogf/gf/v2/util/gconv"
+)
+
+var ModelManage = &modelManageDao{}
+
+type modelManageDao struct{}
+
+// Insert 插入
+func (d *modelManageDao) Insert(ctx context.Context, req *dto.CreateModelManageReq) (id int64, err error) {
+	var e = new(entity.ModelManage)
+	err = gconv.Struct(req, &e)
+	if err != nil {
+		return
+	}
+	r, err := gfdb.DB(ctx, public.DbNameModelGateway).Model(ctx, public.TableNameModelManage).Insert(e)
+	if err != nil {
+		return
+	}
+	return r.LastInsertId()
+}
+
+// Update 更新
+func (d *modelManageDao) Update(ctx context.Context, req *dto.UpdateModelManageReq) (rows int64, err error) {
+	r, err := gfdb.DB(ctx, public.DbNameModelGateway).Model(ctx, public.TableNameModelManage).OmitEmpty().Data(req).Where(entity.ModelManageCol.Id, req.Id).Update()
+	if err != nil {
+		return
+	}
+	return r.RowsAffected()
+}
+
+// Delete 删除
+func (d *modelManageDao) Delete(ctx context.Context, req *dto.DeleteModelManageReq) (rows int64, err error) {
+	r, err := gfdb.DB(ctx, public.DbNameModelGateway).Model(ctx, public.TableNameModelManage).Where(entity.ModelManageCol.Id, req.Id).Delete()
+	if err != nil {
+		return
+	}
+	return r.RowsAffected()
+}
+
+func (d *modelManageDao) Get(ctx context.Context, req *dto.GetModelManage, fields ...string) (res *entity.ModelManage, err error) {
+	r, err := gfdb.DB(ctx, public.DbNameModelGateway).Model(ctx, public.TableNameModelManage).Cache(ctx).
+		OmitEmpty().
+		Where(entity.ModelManageCol.ModelName, req.ModelName).
+		Where(entity.ModelManageCol.ChatModel, req.ChatModel).
+		Where(entity.ModelManageCol.Creator, req.Creator).
+		Fields(fields).One()
+	if err != nil {
+		return
+	}
+	err = r.Struct(&res)
+	return
+}
+
+func (d *modelManageDao) GetNotTenantId(ctx context.Context, req *dto.GetModelManageReq, fields ...string) (res *entity.ModelManage, err error) {
+	// 获取表前缀
+	prefix := g.Cfg().MustGet(ctx, fmt.Sprintf("database.%s.0.prefix", public.DbNameModelGateway)).String()
+	table := prefix + public.TableNameModelManage
+	// 动态拼接 SELECT 列
+	var field string
+	if !g.IsEmpty(fields) {
+		for k, v := range fields {
+			if k == len(fields)-1 {
+				field = field + v
+			} else {
+				field = field + v + ","
+			}
+		}
+	} else {
+		field = "*"
+	}
+	// 动态拼接 WHERE 条件
+	var whereCondition string
+	var queryParams []interface{}
+	if !g.IsEmpty(req.Id) {
+		whereCondition = fmt.Sprintf(" AND %s=(?) ", entity.ModelManageCol.Id)
+		queryParams = append(queryParams, req.Id)
+	}
+	whereCondition = whereCondition + " AND " + entity.ModelManageCol.DeletedAt + " IS NULL "
+
+	sql := `SELECT ` + field + ` FROM ` + table + ` WHERE 1=1 ` + whereCondition + ``
+	// 执行查询
+	result, err := gfdb.DB(ctx, public.DbNameModelGateway).GetOne(ctx, sql, queryParams...)
+	if err != nil {
+		return nil, err
+	}
+	err = result.Struct(&res)
+	return
+}
+func (d *modelManageDao) ListNotTenantId(ctx context.Context, req *dto.ListModelManageReq, fields ...string) (res []*entity.ModelManage, total int, err error) {
+	// 获取表前缀
+	prefix := g.Cfg().MustGet(ctx, fmt.Sprintf("database.%s.0.prefix", public.DbNameModelGateway)).String()
+	table := prefix + public.TableNameModelManage
+
+	// 动态拼接 SELECT 列
+	var field string
+	if !g.IsEmpty(fields) {
+		for k, v := range fields {
+			if k == len(fields)-1 {
+				field = field + v
+			} else {
+				field = field + v + ","
+			}
+		}
+	} else {
+		field = "*"
+	}
+
+	// 动态拼接 WHERE 条件
+	var whereCondition string
+	var queryParams []interface{}
+
+	if !g.IsEmpty(req.ModelName) {
+		whereCondition += fmt.Sprintf(" AND %s=(?) ", entity.ModelManageCol.ModelName)
+		queryParams = append(queryParams, req.ModelName)
+	}
+	if !g.IsEmpty(req.ModelType) {
+		whereCondition += fmt.Sprintf(" AND %s=(?) ", entity.ModelManageCol.ModelType)
+		queryParams = append(queryParams, req.ModelType)
+	}
+	if !g.IsEmpty(req.Creator) {
+		whereCondition += fmt.Sprintf(" AND (%s=(?) OR %s=true) ", entity.ModelManageCol.Creator, entity.ModelManageCol.SystemModel)
+		queryParams = append(queryParams, req.Creator)
+	}
+	whereCondition = whereCondition + " AND " + entity.ModelManageCol.DeletedAt + " IS NULL "
+
+	// 1. 统计去重后总条数
+	countSql := fmt.Sprintf(
+		`SELECT COUNT(DISTINCT %s) FROM %s WHERE 1=1 %s`,
+		entity.ModelManageCol.ModelName,
+		table,
+		whereCondition,
+	)
+	countResult, err := gfdb.DB(ctx, public.DbNameModelGateway).GetOne(ctx, countSql, queryParams...)
+	if err != nil {
+		return nil, 0, err
+	}
+	type crr struct {
+		Count int64 `db:"count"`
+	}
+	var cr crr
+	if err = countResult.Struct(&cr); err != nil {
+		return nil, 0, err
+	}
+	total = int(cr.Count)
+	// 2. 分页处理
+	limitSql := ""
+	if req.Page != nil {
+		pageNum := int(req.Page.PageNum)
+		pageSize := int(req.Page.PageSize)
+		offset := (pageNum - 1) * pageSize
+		limitSql = fmt.Sprintf(" LIMIT ? OFFSET ? ")
+		// PG 语法 LIMIT 条数 OFFSET 偏移量
+		queryParams = append(queryParams, pageSize, offset)
+	}
+
+	// 排序优先级：1.分组字段ModelName  2.SystemModel升序(false在前，保留用户数据)  3.创建时间倒序
+	orderSql := fmt.Sprintf(
+		" ORDER BY %s, %s ASC, %s DESC ",
+		entity.ModelManageCol.ModelName,
+		entity.ModelManageCol.SystemModel,
+		entity.ModelManageCol.CreatedAt,
+	)
+
+	// PG DISTINCT ON 按模型名去重，同名只取第一条(用户数据)
+	sql := fmt.Sprintf(
+		`SELECT DISTINCT ON (%s) %s FROM %s WHERE 1=1 %s %s %s`,
+		entity.ModelManageCol.ModelName,
+		field,
+		table,
+		whereCondition,
+		orderSql,
+		limitSql,
+	)
+
+	// 执行查询
+	result, err := gfdb.DB(ctx, public.DbNameModelGateway).GetAll(ctx, sql, queryParams...)
+	if err != nil {
+		return nil, 0, err
+	}
+	err = result.Structs(&res)
+	return
+}

go.mod

@@ -3,7 +3,7 @@ module model-gateway
 go 1.26.1
 
 require (
-	gitea.redpowerfuture.com/red-future/common v0.0.23
+	gitea.redpowerfuture.com/red-future/common v0.0.30
 	github.com/gogf/gf/contrib/drivers/pgsql/v2 v2.10.2
 	github.com/gogf/gf/contrib/nosql/redis/v2 v2.10.2
 	github.com/gogf/gf/v2 v2.10.2
@@ -12,22 +12,25 @@ require (
 )
 
 require (
-	github.com/BurntSushi/toml v1.5.0 // indirect
+	github.com/BurntSushi/toml v1.6.0 // indirect
 	github.com/armon/go-metrics v0.4.1 // indirect
 	github.com/bwmarrin/snowflake v0.3.0 // indirect
 	github.com/cenkalti/backoff/v5 v5.0.3 // indirect
 	github.com/cespare/xxhash/v2 v2.3.0 // indirect
 	github.com/clbanning/mxj/v2 v2.7.0 // indirect
+	github.com/clipperhouse/displaywidth v0.11.0 // indirect
+	github.com/clipperhouse/uax29/v2 v2.7.0 // indirect
 	github.com/dgraph-io/badger/v4 v4.2.0 // indirect
 	github.com/dgraph-io/ristretto v0.1.1 // indirect
-	github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f // indirect
 	github.com/dustin/go-humanize v1.0.1 // indirect
 	github.com/emirpasic/gods/v2 v2.0.0-alpha // indirect
-	github.com/fatih/color v1.18.0 // indirect
-	github.com/fsnotify/fsnotify v1.9.0 // indirect
+	github.com/fatih/color v1.19.0 // indirect
+	github.com/fsnotify/fsnotify v1.10.1 // indirect
 	github.com/go-ego/gse v1.0.2 // indirect
 	github.com/go-logr/logr v1.4.3 // indirect
 	github.com/go-logr/stdr v1.2.2 // indirect
+	github.com/go-viper/mapstructure/v2 v2.5.0 // indirect
+	github.com/goccy/go-json v0.10.6 // indirect
 	github.com/gogf/gf/contrib/registry/consul/v2 v2.9.5 // indirect
 	github.com/gogf/gf/contrib/trace/otlphttp/v2 v2.9.5 // indirect
 	github.com/gogo/protobuf v1.3.2 // indirect
@@ -36,57 +39,62 @@ require (
 	github.com/golang/groupcache v0.0.0-20210331224755-41bb18bfe9da // indirect
 	github.com/golang/protobuf v1.5.4 // indirect
 	github.com/golang/snappy v1.0.0 // indirect
-	github.com/google/flatbuffers v1.12.1 // indirect
+	github.com/google/btree v1.1.3 // indirect
+	github.com/google/flatbuffers v25.12.19+incompatible // indirect
 	github.com/gorilla/websocket v1.5.4-0.20250319132907-e064f32e3674 // indirect
 	github.com/grokify/html-strip-tags-go v0.1.0 // indirect
-	github.com/grpc-ecosystem/grpc-gateway/v2 v2.27.2 // indirect
-	github.com/hashicorp/consul/api v1.26.1 // indirect
+	github.com/grpc-ecosystem/grpc-gateway/v2 v2.28.0 // indirect
+	github.com/hashicorp/consul/api v1.33.5 // indirect
 	github.com/hashicorp/errwrap v1.1.0 // indirect
 	github.com/hashicorp/go-cleanhttp v0.5.2 // indirect
 	github.com/hashicorp/go-hclog v1.5.0 // indirect
 	github.com/hashicorp/go-immutable-radix v1.3.1 // indirect
 	github.com/hashicorp/go-multierror v1.1.1 // indirect
 	github.com/hashicorp/go-rootcerts v1.0.2 // indirect
+	github.com/hashicorp/go-version v1.9.0 // indirect
 	github.com/hashicorp/golang-lru v1.0.2 // indirect
 	github.com/hashicorp/serf v0.10.1 // indirect
-	github.com/klauspost/compress v1.18.0 // indirect
-	github.com/lib/pq v1.10.9 // indirect
+	github.com/klauspost/compress v1.18.6 // indirect
+	github.com/klauspost/cpuid/v2 v2.3.0 // indirect
+	github.com/lib/pq v1.12.3 // indirect
 	github.com/magiconair/properties v1.8.10 // indirect
-	github.com/mattn/go-colorable v0.1.14 // indirect
-	github.com/mattn/go-isatty v0.0.20 // indirect
-	github.com/mattn/go-runewidth v0.0.16 // indirect
+	github.com/mattn/go-colorable v0.1.15 // indirect
+	github.com/mattn/go-isatty v0.0.22 // indirect
+	github.com/mattn/go-runewidth v0.0.24 // indirect
+	github.com/miekg/dns v1.1.72 // indirect
 	github.com/mitchellh/go-homedir v1.1.0 // indirect
-	github.com/mitchellh/mapstructure v1.5.0 // indirect
-	github.com/olekukonko/errors v1.1.0 // indirect
-	github.com/olekukonko/ll v0.0.9 // indirect
-	github.com/olekukonko/tablewriter v1.1.0 // indirect
+	github.com/olekukonko/cat v0.0.0-20250911104152-50322a0618f6 // indirect
+	github.com/olekukonko/errors v1.3.0 // indirect
+	github.com/olekukonko/ll v0.1.8 // indirect
+	github.com/olekukonko/tablewriter v1.1.4 // indirect
 	github.com/pkg/errors v0.9.1 // indirect
 	github.com/r3labs/diff/v2 v2.15.1 // indirect
-	github.com/redis/go-redis/v9 v9.12.1 // indirect
-	github.com/rivo/uniseg v0.4.7 // indirect
-	github.com/tidwall/match v1.1.1 // indirect
-	github.com/tidwall/pretty v1.2.0 // indirect
+	github.com/redis/go-redis/v9 v9.21.0 // indirect
+	github.com/stretchr/objx v0.5.2 // indirect
+	github.com/tidwall/match v1.2.0 // indirect
+	github.com/tidwall/pretty v1.2.1 // indirect
 	github.com/tiger1103/gfast-token v1.0.10 // indirect
 	github.com/vcaesar/cedar v0.30.0 // indirect
 	github.com/vmihailenco/msgpack v4.0.4+incompatible // indirect
-	go.mongodb.org/mongo-driver/v2 v2.4.0 // indirect
-	go.opencensus.io v0.23.0 // indirect
-	go.opentelemetry.io/auto/sdk v1.1.0 // indirect
-	go.opentelemetry.io/otel v1.38.0 // indirect
+	go.mongodb.org/mongo-driver/v2 v2.6.0 // indirect
+	go.opencensus.io v0.24.0 // indirect
+	go.opentelemetry.io/auto/sdk v1.2.1 // indirect
+	go.opentelemetry.io/otel v1.44.0 // indirect
 	go.opentelemetry.io/otel/exporters/otlp/otlptrace v1.38.0 // indirect
 	go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.38.0 // indirect
-	go.opentelemetry.io/otel/metric v1.38.0 // indirect
-	go.opentelemetry.io/otel/sdk v1.38.0 // indirect
-	go.opentelemetry.io/otel/trace v1.38.0 // indirect
+	go.opentelemetry.io/otel/metric v1.44.0 // indirect
+	go.opentelemetry.io/otel/sdk v1.44.0 // indirect
+	go.opentelemetry.io/otel/trace v1.44.0 // indirect
 	go.opentelemetry.io/proto/otlp v1.7.1 // indirect
-	golang.org/x/exp v0.0.0-20250305212735-054e65f0b394 // indirect
-	golang.org/x/net v0.47.0 // indirect
-	golang.org/x/sys v0.38.0 // indirect
-	golang.org/x/text v0.31.0 // indirect
+	go.uber.org/atomic v1.11.0 // indirect
+	golang.org/x/exp v0.0.0-20260611194520-c48552f49976 // indirect
+	golang.org/x/net v0.56.0 // indirect
+	golang.org/x/sys v0.46.0 // indirect
+	golang.org/x/text v0.38.0 // indirect
 	google.golang.org/appengine v1.6.7 // indirect
-	google.golang.org/genproto/googleapis/api v0.0.0-20250825161204-c5933d9347a5 // indirect
-	google.golang.org/genproto/googleapis/rpc v0.0.0-20250825161204-c5933d9347a5 // indirect
-	google.golang.org/grpc v1.75.0 // indirect
-	google.golang.org/protobuf v1.36.8 // indirect
+	google.golang.org/genproto/googleapis/api v0.0.0-20260209200024-4cfbd4190f57 // indirect
+	google.golang.org/genproto/googleapis/rpc v0.0.0-20260209200024-4cfbd4190f57 // indirect
+	google.golang.org/grpc v1.79.3 // indirect
+	google.golang.org/protobuf v1.36.11 // indirect
 	gopkg.in/yaml.v3 v3.0.1 // indirect
 )

go.sum

@@ -1,9 +1,11 @@
 cloud.google.com/go v0.26.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
-gitea.redpowerfuture.com/red-future/common v0.0.23 h1:xieoA00iKOCDm5SO9iXn+cSyMKBAlZwI0fuEVPWrHLg=
-gitea.redpowerfuture.com/red-future/common v0.0.23/go.mod h1:50U1Xi+Ie56z09S5LQbZvaken0Mxv3OeS9LgR7U/ZRY=
+gitea.redpowerfuture.com/red-future/common v0.0.29 h1:5McaN5pSewvrLUHQzWMX6EaUvD+B5I5bMYoU+clHJk4=
+gitea.redpowerfuture.com/red-future/common v0.0.29/go.mod h1:50U1Xi+Ie56z09S5LQbZvaken0Mxv3OeS9LgR7U/ZRY=
+gitea.redpowerfuture.com/red-future/common v0.0.30 h1:UkWYubUsLPJQUhEhc9Ca2UPg5iLC6jzURo3ngztINYg=
+gitea.redpowerfuture.com/red-future/common v0.0.30/go.mod h1:zuhqbWHd/YICalYJnmecY8Vqo5j4dtZOD63P96fiFeU=
 github.com/BurntSushi/toml v0.3.1/go.mod h1:xHWCNGjB5oqiDr8zfno3MHue2Ht5sIBksp03qcyfWMU=
-github.com/BurntSushi/toml v1.5.0 h1:W5quZX/G/csjUnuI8SUYlsHs9M38FC7znL0lIO+DvMg=
-github.com/BurntSushi/toml v1.5.0/go.mod h1:ukJfTF/6rtPPRCnwkur4qwRxa8vTRFBF0uk2lLoLwho=
+github.com/BurntSushi/toml v1.6.0 h1:dRaEfpa2VI55EwlIW72hMRHdWouJeRF7TPYhI+AUQjk=
+github.com/BurntSushi/toml v1.6.0/go.mod h1:ukJfTF/6rtPPRCnwkur4qwRxa8vTRFBF0uk2lLoLwho=
 github.com/DataDog/datadog-go v3.2.0+incompatible/go.mod h1:LButxg5PwREeZtORoXG3tL4fMGNddJ+vMq1mwgfaqoQ=
 github.com/alecthomas/template v0.0.0-20160405071501-a0175ee3bccc/go.mod h1:LOuyumcjzFXgccqObfd/Ljyb9UuFJ6TxHnclSeseNhc=
 github.com/alecthomas/template v0.0.0-20190718012654-fb15b899a751/go.mod h1:LOuyumcjzFXgccqObfd/Ljyb9UuFJ6TxHnclSeseNhc=
@@ -36,6 +38,10 @@ github.com/circonus-labs/circonusllhist v0.1.3/go.mod h1:kMXHVDlOchFAehlya5ePtbp
 github.com/clbanning/mxj/v2 v2.7.0 h1:WA/La7UGCanFe5NpHF0Q3DNtnCsVoxbPKuyBNHWRyME=
 github.com/clbanning/mxj/v2 v2.7.0/go.mod h1:hNiWqW14h+kc+MdF9C6/YoRfjEJoR3ou6tn/Qo+ve2s=
 github.com/client9/misspell v0.3.4/go.mod h1:qj6jICC3Q7zFZvVWo7KLAzC3yx5G7kyvSDkc90ppPyw=
+github.com/clipperhouse/displaywidth v0.11.0 h1:lBc6kY44VFw+TDx4I8opi/EtL9m20WSEFgwIwO+UVM8=
+github.com/clipperhouse/displaywidth v0.11.0/go.mod h1:bkrFNkf81G8HyVqmKGxsPufD3JhNl3dSqnGhOoSD/o0=
+github.com/clipperhouse/uax29/v2 v2.7.0 h1:+gs4oBZ2gPfVrKPthwbMzWZDaAFPGYK72F0NJv2v7Vk=
+github.com/clipperhouse/uax29/v2 v2.7.0/go.mod h1:EFJ2TJMRUaplDxHKj1qAEhCtQPW2tJSwu5BF98AuoVM=
 github.com/cncf/udpa/go v0.0.0-20191209042840-269d4d468f6f/go.mod h1:M8M6+tZqaGXZJjfX53e64911xZQV5JYwmTeXPW+k8Sc=
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
@@ -47,8 +53,6 @@ github.com/dgraph-io/ristretto v0.1.1 h1:6CWw5tJNgpegArSHpNHJKldNeq03FQCwYvfMVWa
 github.com/dgraph-io/ristretto v0.1.1/go.mod h1:S1GPSBCYCIhmVNfcth17y2zZtQT6wzkzgwUve0VDWWA=
 github.com/dgryski/go-farm v0.0.0-20190423205320-6a90982ecee2 h1:tdlZCpZ/P9DhczCTSixgIKmwPv6+wP5DGjqLYw5SUiA=
 github.com/dgryski/go-farm v0.0.0-20190423205320-6a90982ecee2/go.mod h1:SqUrOPUnsFjfmXRMNPybcSiG0BgUW2AuFH8PAnS2iTw=
-github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f h1:lO4WD4F/rVNCu3HqELle0jiPLLBs70cWOduZpkS1E78=
-github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f/go.mod h1:cuUVRXasLTGF7a8hSLbxyZXjz+1KgoB3wDUb6vlszIc=
 github.com/dustin/go-humanize v1.0.0/go.mod h1:HtrtbFcZ19U5GC7JDqmcUSB87Iq5E25KnS6fMYU6eOk=
 github.com/dustin/go-humanize v1.0.1 h1:GzkhY7T5VNhEkwH0PVJgjz+fX1rhBrR7pRT3mDkpeCY=
 github.com/dustin/go-humanize v1.0.1/go.mod h1:Mu1zIs6XwVuF/gI1OepvI0qD18qycQx+mFykh5fBlto=
@@ -61,10 +65,10 @@ github.com/envoyproxy/protoc-gen-validate v0.1.0/go.mod h1:iSmxcyjqTsJpI2R4NaDN7
 github.com/fatih/color v1.7.0/go.mod h1:Zm6kSWBoL9eyXnKyktHP6abPY2pDugNf5KwzbycvMj4=
 github.com/fatih/color v1.9.0/go.mod h1:eQcE1qtQxscV5RaZvpXrrb8Drkc3/DdQ+uUYCNjL+zU=
 github.com/fatih/color v1.13.0/go.mod h1:kLAiJbzzSOZDVNGyDpeOxJ47H46qBXwg5ILebYFFOfk=
-github.com/fatih/color v1.18.0 h1:S8gINlzdQ840/4pfAwic/ZE0djQEH3wM94VfqLTZcOM=
-github.com/fatih/color v1.18.0/go.mod h1:4FelSpRwEGDpQ12mAdzqdOukCy4u8WUtOY6lkT/6HfU=
-github.com/fsnotify/fsnotify v1.9.0 h1:2Ml+OJNzbYCTzsxtv8vKSFD9PbJjmhYF14k/jKC7S9k=
-github.com/fsnotify/fsnotify v1.9.0/go.mod h1:8jBTzvmWwFyi3Pb8djgCCO5IBqzKJ/Jwo8TRcHyHii0=
+github.com/fatih/color v1.19.0 h1:Zp3PiM21/9Ld6FzSKyL5c/BULoe/ONr9KlbYVOfG8+w=
+github.com/fatih/color v1.19.0/go.mod h1:zNk67I0ZUT1bEGsSGyCZYZNrHuTkJJB+r6Q9VuMi0LE=
+github.com/fsnotify/fsnotify v1.10.1 h1:b0/UzAf9yR5rhf3RPm9gf3ehBPpf0oZKIjtpKrx59Ho=
+github.com/fsnotify/fsnotify v1.10.1/go.mod h1:TLheqan6HD6GBK6PrDWyDPBaEV8LspOxvPSjC+bVfgo=
 github.com/go-ego/gse v1.0.2 h1:+27lYFPhQEhA9igtdOsJPRKYL/k3TwYsxBF5jr6KFv4=
 github.com/go-ego/gse v1.0.2/go.mod h1:Fy35G+q7VV7Et1zIKO8o/sW1kkugV3znXap/lF/11zc=
 github.com/go-kit/kit v0.8.0/go.mod h1:xBxKIO96dXMWWy0MnWVtmwkA9/13aqxPnvrjFYMA2as=
@@ -77,6 +81,10 @@ github.com/go-logr/logr v1.4.3/go.mod h1:9T104GzyrTigFIr8wt5mBrctHMim0Nb2HLGrmQ4
 github.com/go-logr/stdr v1.2.2 h1:hSWxHoqTgW2S2qGc0LTAI563KZ5YKYRhT3MFKZMbjag=
 github.com/go-logr/stdr v1.2.2/go.mod h1:mMo/vtBO5dYbehREoey6XUKy/eSumjCCveDpRre4VKE=
 github.com/go-stack/stack v1.8.0/go.mod h1:v0f6uXyyMGvRgIKkXu+yp6POWl0qKG85gN/melR3HDY=
+github.com/go-viper/mapstructure/v2 v2.5.0 h1:vM5IJoUAy3d7zRSVtIwQgBj7BiWtMPfmPEgAXnvj1Ro=
+github.com/go-viper/mapstructure/v2 v2.5.0/go.mod h1:oJDH3BJKyqBA2TXFhDsKDGDTlndYOZ6rGS0BRZIxGhM=
+github.com/goccy/go-json v0.10.6 h1:p8HrPJzOakx/mn/bQtjgNjdTcN+/S6FcG2CTtQOrHVU=
+github.com/goccy/go-json v0.10.6/go.mod h1:oq7eo15ShAhp70Anwd5lgX2pLfOS3QCiwU/PULtXL6M=
 github.com/gogf/gf/contrib/drivers/pgsql/v2 v2.10.2 h1:u8EpP24GkprogROnJ7htMov9Fc66pTP1eVYrWxiCYOs=
 github.com/gogf/gf/contrib/drivers/pgsql/v2 v2.10.2/go.mod h1:GmvM3r8GVByVMi4RD2+MCs5+CfxVXPMeT8mVDkAaAXE=
 github.com/gogf/gf/contrib/nosql/redis/v2 v2.10.2 h1:iTQegT+lEg/wDKvj2mi3W1wrdrwFarjokf88EXVVgu4=
@@ -114,10 +122,10 @@ github.com/golang/protobuf v1.5.4/go.mod h1:lnTiLA8Wa4RWRcIUkrtSVa5nRhsEGBg48fD6
 github.com/golang/snappy v1.0.0 h1:Oy607GVXHs7RtbggtPBnr2RmDArIsAefDwvrdWvRhGs=
 github.com/golang/snappy v1.0.0/go.mod h1:/XxbfmMg8lxefKM7IXC3fBNl/7bRcc72aCRzEWrmP2Q=
 github.com/google/btree v0.0.0-20180813153112-4030bb1f1f0c/go.mod h1:lNA+9X1NB3Zf8V7Ke586lFgjr2dZNuvo3lPJSGZ5JPQ=
-github.com/google/btree v1.0.1 h1:gK4Kx5IaGY9CD5sPJ36FHiBJ6ZXl0kilRiiCj+jdYp4=
-github.com/google/btree v1.0.1/go.mod h1:xXMiIv4Fb/0kKde4SpL7qlzvu5cMJDRkFDxJfI9uaxA=
-github.com/google/flatbuffers v1.12.1 h1:MVlul7pQNoDzWRLTw5imwYsl+usrS1TXG2H4jg6ImGw=
-github.com/google/flatbuffers v1.12.1/go.mod h1:1AeVuKshWv4vARoZatz6mlQ0JxURH0Kv5+zNeJKJCa8=
+github.com/google/btree v1.1.3 h1:CVpQJjYgC4VbzxeGVHfvZrv1ctoYCAI8vbl07Fcxlyg=
+github.com/google/btree v1.1.3/go.mod h1:qOPhT0dTNdNzV6Z/lhRX0YXUafgPLFUh+gZMl761Gm4=
+github.com/google/flatbuffers v25.12.19+incompatible h1:haMV2JRRJCe1998HeW/p0X9UaMTK6SDo0ffLn2+DbLs=
+github.com/google/flatbuffers v25.12.19+incompatible/go.mod h1:1AeVuKshWv4vARoZatz6mlQ0JxURH0Kv5+zNeJKJCa8=
 github.com/google/go-cmp v0.2.0/go.mod h1:oXzfMopK8JAjlY9xF4vHSVASa0yLyX7SntLO5aqRK0M=
 github.com/google/go-cmp v0.3.0/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU=
 github.com/google/go-cmp v0.3.1/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU=
@@ -134,12 +142,12 @@ github.com/gorilla/websocket v1.5.4-0.20250319132907-e064f32e3674 h1:JeSE6pjso5T
 github.com/gorilla/websocket v1.5.4-0.20250319132907-e064f32e3674/go.mod h1:r4w70xmWCQKmi1ONH4KIaBptdivuRPyosB9RmPlGEwA=
 github.com/grokify/html-strip-tags-go v0.1.0 h1:03UrQLjAny8xci+R+qjCce/MYnpNXCtgzltlQbOBae4=
 github.com/grokify/html-strip-tags-go v0.1.0/go.mod h1:ZdzgfHEzAfz9X6Xe5eBLVblWIxXfYSQ40S/VKrAOGpc=
-github.com/grpc-ecosystem/grpc-gateway/v2 v2.27.2 h1:8Tjv8EJ+pM1xP8mK6egEbD1OgnVTyacbefKhmbLhIhU=
-github.com/grpc-ecosystem/grpc-gateway/v2 v2.27.2/go.mod h1:pkJQ2tZHJ0aFOVEEot6oZmaVEZcRme73eIFmhiVuRWs=
-github.com/hashicorp/consul/api v1.26.1 h1:5oSXOO5fboPZeW5SN+TdGFP/BILDgBm19OrPZ/pICIM=
-github.com/hashicorp/consul/api v1.26.1/go.mod h1:B4sQTeaSO16NtynqrAdwOlahJ7IUDZM9cj2420xYL8A=
-github.com/hashicorp/consul/sdk v0.15.0 h1:2qK9nDrr4tiJKRoxPGhm6B7xJjLVIQqkjiab2M4aKjU=
-github.com/hashicorp/consul/sdk v0.15.0/go.mod h1:r/OmRRPbHOe0yxNahLw7G9x5WG17E1BIECMtCjcPSNo=
+github.com/grpc-ecosystem/grpc-gateway/v2 v2.28.0 h1:HWRh5R2+9EifMyIHV7ZV+MIZqgz+PMpZ14Jynv3O2Zs=
+github.com/grpc-ecosystem/grpc-gateway/v2 v2.28.0/go.mod h1:JfhWUomR1baixubs02l85lZYYOm7LV6om4ceouMv45c=
+github.com/hashicorp/consul/api v1.33.5 h1:Nn6q87zudRU1rLBTJEgaWxz9STCNadilLCD7B8OA5aI=
+github.com/hashicorp/consul/api v1.33.5/go.mod h1:pa6fJOSHKLOzNHpUVeqLDtxA5+J1D7NNzLasuk8eRXA=
+github.com/hashicorp/consul/sdk v0.17.3 h1:oZMMxzQGSsiT+ToOH50y3Qcs0nc9Ud+7L5lRx+EmMU0=
+github.com/hashicorp/consul/sdk v0.17.3/go.mod h1:jnOmYjiNfVRpBaujQ1DFFVs0N6g3S1y6wygSjLTzYfc=
 github.com/hashicorp/errwrap v1.0.0/go.mod h1:YH+1FKiLXxHSkmPseP+kNlulaMuP3n2brvKWEqk/Jc4=
 github.com/hashicorp/errwrap v1.1.0 h1:OxrOeh75EUXMY8TBjag2fzXGZ40LB6IKw45YeGUDY2I=
 github.com/hashicorp/errwrap v1.1.0/go.mod h1:YH+1FKiLXxHSkmPseP+kNlulaMuP3n2brvKWEqk/Jc4=
@@ -169,8 +177,8 @@ github.com/hashicorp/go-uuid v1.0.0/go.mod h1:6SBZvOh/SIDV7/2o3Jml5SYk/TvGqwFJ/b
 github.com/hashicorp/go-uuid v1.0.1/go.mod h1:6SBZvOh/SIDV7/2o3Jml5SYk/TvGqwFJ/bN7x4byOro=
 github.com/hashicorp/go-uuid v1.0.3 h1:2gKiV6YVmrJ1i2CKKa9obLvRieoRGviZFL26PcT/Co8=
 github.com/hashicorp/go-uuid v1.0.3/go.mod h1:6SBZvOh/SIDV7/2o3Jml5SYk/TvGqwFJ/bN7x4byOro=
-github.com/hashicorp/go-version v1.2.1 h1:zEfKbn2+PDgroKdiOzqiE8rsmLqU2uwi5PB5pBJ3TkI=
-github.com/hashicorp/go-version v1.2.1/go.mod h1:fltr4n8CU8Ke44wwGCBoEymUuxUHl09ZGVZPK5anwXA=
+github.com/hashicorp/go-version v1.9.0 h1:CeOIz6k+LoN3qX9Z0tyQrPtiB1DFYRPfCIBtaXPSCnA=
+github.com/hashicorp/go-version v1.9.0/go.mod h1:fltr4n8CU8Ke44wwGCBoEymUuxUHl09ZGVZPK5anwXA=
 github.com/hashicorp/golang-lru v0.5.0/go.mod h1:/m3WP610KZHVQ1SGc6re/UDhFvYD7pJ4Ao+sR/qLZy8=
 github.com/hashicorp/golang-lru v1.0.2 h1:dV3g9Z/unq5DpblPpw+Oqcv4dU/1omnb4Ok8iPY6p1c=
 github.com/hashicorp/golang-lru v1.0.2/go.mod h1:iADmTwqILo4mZ8BN3D2Q6+9jd8WM5uGBxy+E8yxSoD4=
@@ -185,8 +193,10 @@ github.com/json-iterator/go v1.1.9/go.mod h1:KdQUCv79m/52Kvf8AW2vK1V8akMuk1QjK/u
 github.com/julienschmidt/httprouter v1.2.0/go.mod h1:SYymIcj16QtmaHHD7aYtjjsJG7VTCxuUUipMqKk8s4w=
 github.com/kisielk/errcheck v1.5.0/go.mod h1:pFxgyoBC7bSaBwPgfKdkLd5X25qrDl4LWUI2bnpBCr8=
 github.com/kisielk/gotool v1.0.0/go.mod h1:XhKaO+MFFWcvkIS/tQcRk01m1F5IRFswLeQ+oQHNcck=
-github.com/klauspost/compress v1.18.0 h1:c/Cqfb0r+Yi+JtIEq73FWXVkRonBlf0CRNYc8Zttxdo=
-github.com/klauspost/compress v1.18.0/go.mod h1:2Pp+KzxcywXVXMr50+X0Q/Lsb43OQHYWRCY2AiWywWQ=
+github.com/klauspost/compress v1.18.6 h1:2jupLlAwFm95+YDR+NwD2MEfFO9d4z4Prjl1XXDjuao=
+github.com/klauspost/compress v1.18.6/go.mod h1:cwPg85FWrGar70rWktvGQj8/hthj3wpl0PGDogxkrSQ=
+github.com/klauspost/cpuid/v2 v2.3.0 h1:S4CRMLnYUhGeDFDqkGriYKdfoFlDnMtqTiI/sFzhA9Y=
+github.com/klauspost/cpuid/v2 v2.3.0/go.mod h1:hqwkgyIinND0mEev00jJYCxPNVRVXFQeu1XKlok6oO0=
 github.com/konsorten/go-windows-terminal-sequences v1.0.1/go.mod h1:T0+1ngSBFLxvqU3pZ+m/2kptfBszLMUkC4ZK/EgS/cQ=
 github.com/kr/logfmt v0.0.0-20140226030751-b84e30acd515/go.mod h1:+0opPa2QZZtGFBFZlji/RkVcI2GknAs/DXo4wKdlNEc=
 github.com/kr/pretty v0.1.0/go.mod h1:dAy3ld7l9f0ibDNOQOHHMYYIIbhfbHSm3C4ZsoJORNo=
@@ -196,8 +206,8 @@ github.com/kr/pty v1.1.1/go.mod h1:pFQYn66WHrOpPYNljwOMqo10TkYh1fy3cYio2l3bCsQ=
 github.com/kr/text v0.1.0/go.mod h1:4Jbv+DJW3UT/LiOwJeYQe1efqtUx/iVham/4vfdArNI=
 github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
 github.com/kr/text v0.2.0/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
-github.com/lib/pq v1.10.9 h1:YXG7RB+JIjhP29X+OtkiDnYaXQwpS4JEWq7dtCCRUEw=
-github.com/lib/pq v1.10.9/go.mod h1:AlVN5x4E4T544tWzH6hKfbfQvm3HdbOxrmggDNAPY9o=
+github.com/lib/pq v1.12.3 h1:tTWxr2YLKwIvK90ZXEw8GP7UFHtcbTtty8zsI+YjrfQ=
+github.com/lib/pq v1.12.3/go.mod h1:/p+8NSbOcwzAEI7wiMXFlgydTwcgTr3OSKMsD2BitpA=
 github.com/magiconair/properties v1.8.10 h1:s31yESBquKXCV9a/ScB3ESkOjUYYv+X0rg8SYxI99mE=
 github.com/magiconair/properties v1.8.10/go.mod h1:Dhd985XPs7jluiymwWYZ0G4Z61jb3vdS329zhj2hYo0=
 github.com/mattn/go-colorable v0.0.9/go.mod h1:9vuHe8Xs5qXnSaW/c/ABM9alt+Vo+STaOChaDxuIBZU=
@@ -205,39 +215,39 @@ github.com/mattn/go-colorable v0.1.4/go.mod h1:U0ppj6V5qS13XJ6of8GYAs25YV2eR4EVc
 github.com/mattn/go-colorable v0.1.6/go.mod h1:u6P/XSegPjTcexA+o6vUJrdnUu04hMope9wVRipJSqc=
 github.com/mattn/go-colorable v0.1.9/go.mod h1:u6P/XSegPjTcexA+o6vUJrdnUu04hMope9wVRipJSqc=
 github.com/mattn/go-colorable v0.1.12/go.mod h1:u5H1YNBxpqRaxsYJYSkiCWKzEfiAb1Gb520KVy5xxl4=
-github.com/mattn/go-colorable v0.1.14 h1:9A9LHSqF/7dyVVX6g0U9cwm9pG3kP9gSzcuIPHPsaIE=
-github.com/mattn/go-colorable v0.1.14/go.mod h1:6LmQG8QLFO4G5z1gPvYEzlUgJ2wF+stgPZH1UqBm1s8=
+github.com/mattn/go-colorable v0.1.15 h1:+u9SLTRGnXv73cEsnsmoZBom+dMU88B2M0aDcWy0/jY=
+github.com/mattn/go-colorable v0.1.15/go.mod h1:6LmQG8QLFO4G5z1gPvYEzlUgJ2wF+stgPZH1UqBm1s8=
 github.com/mattn/go-isatty v0.0.3/go.mod h1:M+lRXTBqGeGNdLjl/ufCoiOlB5xdOkqRJdNxMWT7Zi4=
 github.com/mattn/go-isatty v0.0.8/go.mod h1:Iq45c/XA43vh69/j3iqttzPXn0bhXyGjM0Hdxcsrc5s=
 github.com/mattn/go-isatty v0.0.11/go.mod h1:PhnuNfih5lzO57/f3n+odYbM4JtupLOxQOAqxQCu2WE=
 github.com/mattn/go-isatty v0.0.12/go.mod h1:cbi8OIDigv2wuxKPP5vlRcQ1OAZbq2CE4Kysco4FUpU=
 github.com/mattn/go-isatty v0.0.14/go.mod h1:7GGIvUiUoEMVVmxf/4nioHXj79iQHKdU27kJ6hsGG94=
-github.com/mattn/go-isatty v0.0.20 h1:xfD0iDuEKnDkl03q4limB+vH+GxLEtL/jb4xVJSWWEY=
-github.com/mattn/go-isatty v0.0.20/go.mod h1:W+V8PltTTMOvKvAeJH7IuucS94S2C6jfK/D7dTCTo3Y=
-github.com/mattn/go-runewidth v0.0.16 h1:E5ScNMtiwvlvB5paMFdw9p4kSQzbXFikJ5SQO6TULQc=
-github.com/mattn/go-runewidth v0.0.16/go.mod h1:Jdepj2loyihRzMpdS35Xk/zdY8IAYHsh153qUoGf23w=
+github.com/mattn/go-isatty v0.0.22 h1:j8l17JJ9i6VGPUFUYoTUKPSgKe/83EYU2zBC7YNKMw4=
+github.com/mattn/go-isatty v0.0.22/go.mod h1:ZXfXG4SQHsB/w3ZeOYbR0PrPwLy+n6xiMrJlRFqopa4=
+github.com/mattn/go-runewidth v0.0.24 h1:cpokDiIn0MGnhdHwuWnJBITySJ20QyNGnY2kR/ay2DU=
+github.com/mattn/go-runewidth v0.0.24/go.mod h1:XBkDxAl56ILZc9knddidhrOlY5R/pDhgLpndooCuJAs=
 github.com/matttproud/golang_protobuf_extensions v1.0.1/go.mod h1:D8He9yQNgCq6Z5Ld7szi9bcBfOoFv/3dc6xSMkL2PC0=
 github.com/miekg/dns v1.1.26/go.mod h1:bPDLeHnStXmXAq1m/Ch/hvfNHr14JKNPMBo3VZKjuso=
 github.com/miekg/dns v1.1.41/go.mod h1:p6aan82bvRIyn+zDIv9xYNUpwa73JcSh9BKwknJysuI=
-github.com/miekg/dns v1.1.63 h1:8M5aAw6OMZfFXTT7K5V0Eu5YiiL8l7nUAkyN6C9YwaY=
-github.com/miekg/dns v1.1.63/go.mod h1:6NGHfjhpmr5lt3XPLuyfDJi5AXbNIPM9PY6H6sF1Nfs=
+github.com/miekg/dns v1.1.72 h1:vhmr+TF2A3tuoGNkLDFK9zi36F2LS+hKTRW0Uf8kbzI=
+github.com/miekg/dns v1.1.72/go.mod h1:+EuEPhdHOsfk6Wk5TT2CzssZdqkmFhf8r+aVyDEToIs=
 github.com/mitchellh/cli v1.1.0/go.mod h1:xcISNoH86gajksDmfB23e/pu+B+GeFRMYmoHXxx3xhI=
 github.com/mitchellh/go-homedir v1.1.0 h1:lukF9ziXFxDFPkA1vsr5zpc1XuPDn/wFntq5mG+4E0Y=
 github.com/mitchellh/go-homedir v1.1.0/go.mod h1:SfyaCUpYCn1Vlf4IUYiD9fPX4A5wJrkLzIz1N1q0pr0=
 github.com/mitchellh/mapstructure v0.0.0-20160808181253-ca63d7c062ee/go.mod h1:FVVH3fgwuzCH5S8UJGiWEs2h04kUh9fWfEaFds41c1Y=
-github.com/mitchellh/mapstructure v1.5.0 h1:jeMsZIYE/09sWLaz43PL7Gy6RuMjD2eJVyuac5Z2hdY=
-github.com/mitchellh/mapstructure v1.5.0/go.mod h1:bFUtVrKA4DC2yAKiSyO/QUcy7e+RRV2QTWOzhPopBRo=
 github.com/modern-go/concurrent v0.0.0-20180228061459-e0a39a4cb421/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
 github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
 github.com/modern-go/reflect2 v0.0.0-20180701023420-4b7aa43c6742/go.mod h1:bx2lNnkwVCuqBIxFjflWJWanXIb3RllmbCylyMrvgv0=
 github.com/modern-go/reflect2 v1.0.1/go.mod h1:bx2lNnkwVCuqBIxFjflWJWanXIb3RllmbCylyMrvgv0=
 github.com/mwitkow/go-conntrack v0.0.0-20161129095857-cc309e4a2223/go.mod h1:qRWi+5nqEBWmkhHvq77mSJWrCKwh8bxhgT7d/eI7P4U=
-github.com/olekukonko/errors v1.1.0 h1:RNuGIh15QdDenh+hNvKrJkmxxjV4hcS50Db478Ou5sM=
-github.com/olekukonko/errors v1.1.0/go.mod h1:ppzxA5jBKcO1vIpCXQ9ZqgDh8iwODz6OXIGKU8r5m4Y=
-github.com/olekukonko/ll v0.0.9 h1:Y+1YqDfVkqMWuEQMclsF9HUR5+a82+dxJuL1HHSRpxI=
-github.com/olekukonko/ll v0.0.9/go.mod h1:En+sEW0JNETl26+K8eZ6/W4UQ7CYSrrgg/EdIYT2H8g=
-github.com/olekukonko/tablewriter v1.1.0 h1:N0LHrshF4T39KvI96fn6GT8HEjXRXYNDrDjKFDB7RIY=
-github.com/olekukonko/tablewriter v1.1.0/go.mod h1:5c+EBPeSqvXnLLgkm9isDdzR3wjfBkHR9Nhfp3NWrzo=
+github.com/olekukonko/cat v0.0.0-20250911104152-50322a0618f6 h1:zrbMGy9YXpIeTnGj4EljqMiZsIcE09mmF8XsD5AYOJc=
+github.com/olekukonko/cat v0.0.0-20250911104152-50322a0618f6/go.mod h1:rEKTHC9roVVicUIfZK7DYrdIoM0EOr8mK1Hj5s3JjH0=
+github.com/olekukonko/errors v1.3.0 h1:teJvgLGUEqMzBUms+Dj3/3szNqCG/Jdw9iDbum8fR6U=
+github.com/olekukonko/errors v1.3.0/go.mod h1:ppzxA5jBKcO1vIpCXQ9ZqgDh8iwODz6OXIGKU8r5m4Y=
+github.com/olekukonko/ll v0.1.8 h1:ysHCJRGHYKzmBSdz9w5AySztx7lG8SQY+naTGYUbsz8=
+github.com/olekukonko/ll v0.1.8/go.mod h1:RPRC6UcscfFZgjo1nulkfMH5IM0QAYim0LfnMvUuozw=
+github.com/olekukonko/tablewriter v1.1.4 h1:ORUMI3dXbMnRlRggJX3+q7OzQFDdvgbN9nVWj1drm6I=
+github.com/olekukonko/tablewriter v1.1.4/go.mod h1:+kedxuyTtgoZLwif3P1Em4hARJs+mVnzKxmsCL/C5RY=
 github.com/pascaldekloe/goe v0.0.0-20180627143212-57f6aae5913c/go.mod h1:lzWF7FIEvWOWxwDKqyGYQf6ZUaNfKdP144TG7ZOy1lc=
 github.com/pascaldekloe/goe v0.1.0 h1:cBOtyMzM9HTpWjXfbbunk26uA6nG3a8n06Wieeh0MwY=
 github.com/pascaldekloe/goe v0.1.0/go.mod h1:lzWF7FIEvWOWxwDKqyGYQf6ZUaNfKdP144TG7ZOy1lc=
@@ -264,13 +274,10 @@ github.com/prometheus/procfs v0.0.2/go.mod h1:TjEm7ze935MbeOT/UhFTIMYKhuLP4wbCsT
 github.com/prometheus/procfs v0.0.8/go.mod h1:7Qr8sr6344vo1JqZ6HhLceV9o3AJ1Ff+GxbHq6oeK9A=
 github.com/r3labs/diff/v2 v2.15.1 h1:EOrVqPUzi+njlumoqJwiS/TgGgmZo83619FNDB9xQUg=
 github.com/r3labs/diff/v2 v2.15.1/go.mod h1:I8noH9Fc2fjSaMxqF3G2lhDdC0b+JXCfyx85tWFM9kc=
-github.com/redis/go-redis/v9 v9.12.1 h1:k5iquqv27aBtnTm2tIkROUDp8JBXhXZIVu1InSgvovg=
-github.com/redis/go-redis/v9 v9.12.1/go.mod h1:huWgSWd8mW6+m0VPhJjSSQ+d6Nh1VICQ6Q5lHuCH/Iw=
-github.com/rivo/uniseg v0.2.0/go.mod h1:J6wj4VEh+S6ZtnVlnTBMWIodfgj8LQOQFoIToxlJtxc=
-github.com/rivo/uniseg v0.4.7 h1:WUdvkW8uEhrYfLC4ZzdpI2ztxP1I582+49Oc5Mq64VQ=
-github.com/rivo/uniseg v0.4.7/go.mod h1:FN3SvrM+Zdj16jyLfmOkMNblXMcoc8DfTHruCPUcx88=
-github.com/rogpeppe/go-internal v1.13.1 h1:KvO1DLK/DRN07sQ1LQKScxyZJuNnedQ5/wKSR38lUII=
-github.com/rogpeppe/go-internal v1.13.1/go.mod h1:uMEvuHeurkdAXX61udpOXGD/AzZDWNMNyH2VO9fmH0o=
+github.com/redis/go-redis/v9 v9.21.0 h1:FPBE4hhbAke+TLmcY3WkpbDffJEomdqPn3HYiqAtL9E=
+github.com/redis/go-redis/v9 v9.21.0/go.mod h1:v/M13XI1PVCDcm01VtPFOADfZtHf8YW3baQf57KlIkA=
+github.com/rogpeppe/go-internal v1.14.1 h1:UQB4HGPB6osV0SQTLymcB4TgvyWu6ZyliaW0tI/otEQ=
+github.com/rogpeppe/go-internal v1.14.1/go.mod h1:MaRKkUm5W0goXpeCfT7UZI6fk/L7L7so1lCWt35ZSgc=
 github.com/ryanuber/columnize v0.0.0-20160712163229-9b3edd62028f/go.mod h1:sm1tb6uqfes/u+d4ooFouqFdy9/2g9QGwK3SQygK0Ts=
 github.com/sean-/seed v0.0.0-20170313163322-e2103e2c3529 h1:nn5Wsu0esKSJiIVhscUtVbo7ada43DJhG55ua/hjS5I=
 github.com/sean-/seed v0.0.0-20170313163322-e2103e2c3529/go.mod h1:DxrIzT+xaE7yg65j358z/aeFdxmN0P9QXhEzd20vsDc=
@@ -278,22 +285,26 @@ github.com/sirupsen/logrus v1.2.0/go.mod h1:LxeOpSwHxABJmUn/MG1IvRgCAasNZTLOkJPx
 github.com/sirupsen/logrus v1.4.2/go.mod h1:tLMulIdttU9McNUspp0xgXVQah82FyeX6MwdIuYE2rE=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/objx v0.1.1/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
-github.com/stretchr/objx v0.5.0 h1:1zr/of2m5FGMsad5YfcqgdqdWrIhu+EBEJRhR1U7z/c=
+github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
 github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=
+github.com/stretchr/objx v0.5.2 h1:xuMeJ0Sdp5ZMRXx/aWO6RZxdr3beISkG5/G/aIRr3pY=
+github.com/stretchr/objx v0.5.2/go.mod h1:FRsXN1f5AsAjCGJKqEizvkpNtU+EGNCLh3NxZ/8L+MA=
 github.com/stretchr/testify v1.2.2/go.mod h1:a8OnRcib4nhh0OaRAV+Yts87kKdq0PP7pXfy6kDkUVs=
 github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
 github.com/stretchr/testify v1.4.0/go.mod h1:j7eGeouHqKxXV5pUuKE4zz7dFj8WfuZ+81PSLYec5m4=
 github.com/stretchr/testify v1.5.1/go.mod h1:5W2xD1RspED5o8YsWQXVCued0rvSQ+mT+I5cxcmMvtA=
-github.com/stretchr/testify v1.6.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
+github.com/stretchr/testify v1.7.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
 github.com/stretchr/testify v1.7.2/go.mod h1:R6va5+xMeoiuVRoj+gSkQ7d3FALtqAAGI1FQKckRals=
+github.com/stretchr/testify v1.8.0/go.mod h1:yNjHg4UonilssWZ8iaSj1OCr/vHnekPRkoO+kdMU+MU=
+github.com/stretchr/testify v1.8.1/go.mod h1:w2LPCIKwWwSfY2zedu0+kehJoqGctiVI29o6fzry7u4=
 github.com/stretchr/testify v1.11.1 h1:7s2iGBzp5EwR7/aIZr8ao5+dra3wiQyKjjFuvgVKu7U=
 github.com/stretchr/testify v1.11.1/go.mod h1:wZwfW3scLgRK+23gO65QZefKpKQRnfz6sD981Nm4B6U=
 github.com/tidwall/gjson v1.19.0 h1:xwxm7n691Uf3u5OFjzngavjGTh55KX5q/9w9xHW88JU=
 github.com/tidwall/gjson v1.19.0/go.mod h1:V37/opeE/JbLUOfH0QTXiNez2l0RUjYUhpT4szFQAfc=
-github.com/tidwall/match v1.1.1 h1:+Ho715JplO36QYgwN9PGYNhgZvoUSc9X2c80KVTi+GA=
-github.com/tidwall/match v1.1.1/go.mod h1:eRSPERbgtNPcGhD8UCthc6PmLEQXEWd3PRB5JTxsfmM=
-github.com/tidwall/pretty v1.2.0 h1:RWIZEg2iJ8/g6fDDYzMpobmaoGh5OLl4AXtGUGPcqCs=
-github.com/tidwall/pretty v1.2.0/go.mod h1:ITEVvHYasfjBbM0u2Pg8T2nJnzm8xPwvNhhsoaGGjNU=
+github.com/tidwall/match v1.2.0 h1:0pt8FlkOwjN2fPt4bIl4BoNxb98gGHN2ObFEDkrfZnM=
+github.com/tidwall/match v1.2.0/go.mod h1:eRSPERbgtNPcGhD8UCthc6PmLEQXEWd3PRB5JTxsfmM=
+github.com/tidwall/pretty v1.2.1 h1:qjsOFOWWQl+N3RsoF5/ssm1pHmJJwhjlSbZ51I6wMl4=
+github.com/tidwall/pretty v1.2.1/go.mod h1:ITEVvHYasfjBbM0u2Pg8T2nJnzm8xPwvNhhsoaGGjNU=
 github.com/tiger1103/gfast-token v1.0.10 h1:fNiBE/Dq5iTHvTGlCx3DmXa2o4hr0NtumFpffZ39k6s=
 github.com/tiger1103/gfast-token v1.0.10/go.mod h1:a/21mxmj7zFeNvjhZSC0XpEAFHfb1aT2k6DXnufFU1s=
 github.com/tv42/httpunix v0.0.0-20150427012821-b75d8614f926/go.mod h1:9ESjWnEqriFuLhtthL60Sar/7RFoluCcXsuvEwTV5KM=
@@ -305,28 +316,32 @@ github.com/vmihailenco/msgpack v4.0.4+incompatible h1:dSLoQfGFAo3F6OoNhwUmLwVgaU
 github.com/vmihailenco/msgpack v4.0.4+incompatible/go.mod h1:fy3FlTQTDXWkZ7Bh6AcGMlsjHatGryHQYUTf1ShIgkk=
 github.com/yuin/goldmark v1.1.27/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 github.com/yuin/goldmark v1.2.1/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
-go.mongodb.org/mongo-driver/v2 v2.4.0 h1:Oq6BmUAAFTzMeh6AonuDlgZMuAuEiUxoAD1koK5MuFo=
-go.mongodb.org/mongo-driver/v2 v2.4.0/go.mod h1:jHeEDJHJq7tm6ZF45Issun9dbogjfnPySb1vXA7EeAI=
-go.opencensus.io v0.23.0 h1:gqCw0LfLxScz8irSi8exQc7fyQ0fKQU/qnC/X8+V/1M=
-go.opencensus.io v0.23.0/go.mod h1:XItmlyltB5F7CS4xOC1DcqMoFqwtC6OG2xF7mCv7P7E=
-go.opentelemetry.io/auto/sdk v1.1.0 h1:cH53jehLUN6UFLY71z+NDOiNJqDdPRaXzTel0sJySYA=
-go.opentelemetry.io/auto/sdk v1.1.0/go.mod h1:3wSPjt5PWp2RhlCcmmOial7AvC4DQqZb7a7wCow3W8A=
-go.opentelemetry.io/otel v1.38.0 h1:RkfdswUDRimDg0m2Az18RKOsnI8UDzppJAtj01/Ymk8=
-go.opentelemetry.io/otel v1.38.0/go.mod h1:zcmtmQ1+YmQM9wrNsTGV/q/uyusom3P8RxwExxkZhjM=
+github.com/zeebo/xxh3 v1.1.0 h1:s7DLGDK45Dyfg7++yxI0khrfwq9661w9EN78eP/UZVs=
+github.com/zeebo/xxh3 v1.1.0/go.mod h1:IisAie1LELR4xhVinxWS5+zf1lA4p0MW4T+w+W07F5s=
+go.mongodb.org/mongo-driver/v2 v2.6.0 h1:b9sJOYrkmt4l8bY43ZenFBcPlhYIjaOfYHLtbB/5qi8=
+go.mongodb.org/mongo-driver/v2 v2.6.0/go.mod h1:yOI9kBsufol30iFsl1slpdq1I0eHPzybRWdyYUs8K/0=
+go.opencensus.io v0.24.0 h1:y73uSU6J157QMP2kn2r30vwW1A2W2WFwSCGnAVxeaD0=
+go.opencensus.io v0.24.0/go.mod h1:vNK8G9p7aAivkbmorf4v+7Hgx+Zs0yY+0fOtgBfjQKo=
+go.opentelemetry.io/auto/sdk v1.2.1 h1:jXsnJ4Lmnqd11kwkBV2LgLoFMZKizbCi5fNZ/ipaZ64=
+go.opentelemetry.io/auto/sdk v1.2.1/go.mod h1:KRTj+aOaElaLi+wW1kO/DZRXwkF4C5xPbEe3ZiIhN7Y=
+go.opentelemetry.io/otel v1.44.0 h1:JjwHmHpA4iZ3wBxluu2fbbE7j4kqlE8jXyAyPXH7HqU=
+go.opentelemetry.io/otel v1.44.0/go.mod h1:BMgjTHL9WPRlRjL2oZCBTL4whCGtXch2H4BhOPIAyYc=
 go.opentelemetry.io/otel/exporters/otlp/otlptrace v1.38.0 h1:GqRJVj7UmLjCVyVJ3ZFLdPRmhDUp2zFmQe3RHIOsw24=
 go.opentelemetry.io/otel/exporters/otlp/otlptrace v1.38.0/go.mod h1:ri3aaHSmCTVYu2AWv44YMauwAQc0aqI9gHKIcSbI1pU=
 go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.38.0 h1:aTL7F04bJHUlztTsNGJ2l+6he8c+y/b//eR0jjjemT4=
 go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.38.0/go.mod h1:kldtb7jDTeol0l3ewcmd8SDvx3EmIE7lyvqbasU3QC4=
-go.opentelemetry.io/otel/metric v1.38.0 h1:Kl6lzIYGAh5M159u9NgiRkmoMKjvbsKtYRwgfrA6WpA=
-go.opentelemetry.io/otel/metric v1.38.0/go.mod h1:kB5n/QoRM8YwmUahxvI3bO34eVtQf2i4utNVLr9gEmI=
-go.opentelemetry.io/otel/sdk v1.38.0 h1:l48sr5YbNf2hpCUj/FoGhW9yDkl+Ma+LrVl8qaM5b+E=
-go.opentelemetry.io/otel/sdk v1.38.0/go.mod h1:ghmNdGlVemJI3+ZB5iDEuk4bWA3GkTpW+DOoZMYBVVg=
-go.opentelemetry.io/otel/sdk/metric v1.38.0 h1:aSH66iL0aZqo//xXzQLYozmWrXxyFkBJ6qT5wthqPoM=
-go.opentelemetry.io/otel/sdk/metric v1.38.0/go.mod h1:dg9PBnW9XdQ1Hd6ZnRz689CbtrUp0wMMs9iPcgT9EZA=
-go.opentelemetry.io/otel/trace v1.38.0 h1:Fxk5bKrDZJUH+AMyyIXGcFAPah0oRcT+LuNtJrmcNLE=
-go.opentelemetry.io/otel/trace v1.38.0/go.mod h1:j1P9ivuFsTceSWe1oY+EeW3sc+Pp42sO++GHkg4wwhs=
+go.opentelemetry.io/otel/metric v1.44.0 h1:1w0gILTcHdr3YI+ixLyjemwrVnsMURbTZFrSYCdDdmc=
+go.opentelemetry.io/otel/metric v1.44.0/go.mod h1:8O7hanEPBNgEMmybD3s2VBKcgWOCsA6tzHBPODAiquo=
+go.opentelemetry.io/otel/sdk v1.44.0 h1:nHYwb9lK+fJPU/dnT6s7W7Z8itMWyqrnVfbheVYrZ58=
+go.opentelemetry.io/otel/sdk v1.44.0/go.mod h1:Osuydd3Se74nqjAKxid74N5eC+jfEqfTegHRnq58oK0=
+go.opentelemetry.io/otel/sdk/metric v1.44.0 h1:3LlKgI+VjbVsjNRFZJZAJ30WjXC5VkNRks6si09iEfI=
+go.opentelemetry.io/otel/sdk/metric v1.44.0/go.mod h1:5B5pMARnXxKhltooO4xUuCBorl65a4EpnTalObqOigA=
+go.opentelemetry.io/otel/trace v1.44.0 h1:jxF5CsGYCe74MCRx2X4g7WsY/VBKRqqpNvXlX/6gtIk=
+go.opentelemetry.io/otel/trace v1.44.0/go.mod h1:oLl1jrMQAVo6v3GAggN+1VH9VIz9iUSvW53sW1Q8PIE=
 go.opentelemetry.io/proto/otlp v1.7.1 h1:gTOMpGDb0WTBOP8JaO72iL3auEZhVmAQg4ipjOVAtj4=
 go.opentelemetry.io/proto/otlp v1.7.1/go.mod h1:b2rVh6rfI/s2pHWNlB7ILJcRALpcNDzKhACevjI+ZnE=
+go.uber.org/atomic v1.11.0 h1:ZvwS0R+56ePWxUNi+Atn9dWONBPp/AUETXlHW0DxSjE=
+go.uber.org/atomic v1.11.0/go.mod h1:LUxbIzbOniOlMKjJjyPfpl4v+PKK2cNJn91OQbhoJI0=
 go.uber.org/goleak v1.3.0 h1:2K3zAYmnTNqV73imy9J1T3WC+gmCePx2hEGkimedGto=
 go.uber.org/goleak v1.3.0/go.mod h1:CoHD4mav9JJNrW/WLlf7HGZPjdw8EucARQHekz1X6bE=
 golang.org/x/crypto v0.0.0-20180904163835-0709b304e793/go.mod h1:6SG95UA2DQfeDnfUPMdvaQW0Q7yPrPDi9nlGo2tz2b4=
@@ -335,15 +350,15 @@ golang.org/x/crypto v0.0.0-20190923035154-9ee001bba392/go.mod h1:/lpIB1dKB+9EgE3
 golang.org/x/crypto v0.0.0-20191011191535-87dc89f01550/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
 golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9/go.mod h1:LzIPMQfyMNhhGPhUkYOs5KpL4U8rLKemX1yGLhDgUto=
 golang.org/x/exp v0.0.0-20190121172915-509febef88a4/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
-golang.org/x/exp v0.0.0-20250305212735-054e65f0b394 h1:nDVHiLt8aIbd/VzvPWN6kSOPE7+F/fNFDSXLVYkE/Iw=
-golang.org/x/exp v0.0.0-20250305212735-054e65f0b394/go.mod h1:sIifuuw/Yco/y6yb6+bDNfyeQ/MdPUy/hKEMYQV17cM=
+golang.org/x/exp v0.0.0-20260611194520-c48552f49976 h1:X8Hz2ImujgbmetVuW+w2YkyZChE3cBpZi2P158rTG9M=
+golang.org/x/exp v0.0.0-20260611194520-c48552f49976/go.mod h1:vnf4pv9iKZXY58sQE1L86zmNWJ4159e1RkcWiLCkeEY=
 golang.org/x/lint v0.0.0-20181026193005-c67002cb31c3/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE=
 golang.org/x/lint v0.0.0-20190227174305-5b3e6a55c961/go.mod h1:wehouNa3lNwaWXcvxsM5YxQ5yQlVC4a0KAMCusXpPoU=
 golang.org/x/lint v0.0.0-20190313153728-d0100b6bd8b3/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc=
 golang.org/x/mod v0.2.0/go.mod h1:s0Qsj1ACt9ePp/hMypM3fl4fZqREWJwdYDEqhRiZZUA=
 golang.org/x/mod v0.3.0/go.mod h1:s0Qsj1ACt9ePp/hMypM3fl4fZqREWJwdYDEqhRiZZUA=
-golang.org/x/mod v0.29.0 h1:HV8lRxZC4l2cr3Zq1LvtOsi/ThTgWnUk/y64QSs8GwA=
-golang.org/x/mod v0.29.0/go.mod h1:NyhrlYXJ2H4eJiRy/WDBO6HMqZQ6q9nk4JzS3NuCK+w=
+golang.org/x/mod v0.37.0 h1:vF1DjpVEshcIqoEaauuHebaLk1O1forxjxBaVn884JQ=
+golang.org/x/mod v0.37.0/go.mod h1:m8S8VeM9r4dzDwjrKO0a1sZP3YjeMamRRlD+fmR2Q/0=
 golang.org/x/net v0.0.0-20180724234803-3673e40ba225/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
 golang.org/x/net v0.0.0-20180826012351-8a410e7b638d/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
 golang.org/x/net v0.0.0-20181114220301-adae6a3d119a/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
@@ -359,8 +374,8 @@ golang.org/x/net v0.0.0-20201021035429-f5854403a974/go.mod h1:sp8m0HH+o8qH0wwXwY
 golang.org/x/net v0.0.0-20201110031124-69a78807bb2b/go.mod h1:sp8m0HH+o8qH0wwXwYZr8TS3Oi6o0r6Gce1SSxlDquU=
 golang.org/x/net v0.0.0-20210226172049-e18ecbb05110/go.mod h1:m0MpNAwzfU5UDzcl9v0D8zg8gWTRqZa9RBIspLL5mdg=
 golang.org/x/net v0.0.0-20210410081132-afb366fc7cd1/go.mod h1:9tjilg8BloeKEkVJvy7fQ90B1CfIiPueXVOjqfkSzI8=
-golang.org/x/net v0.47.0 h1:Mx+4dIFzqraBXUugkia1OOvlD6LemFo1ALMHjrXDOhY=
-golang.org/x/net v0.47.0/go.mod h1:/jNxtkgq5yWUGYkaZGqo27cfGZ1c5Nen03aYrrKpVRU=
+golang.org/x/net v0.56.0 h1:Rw8j/hFzGvJUZwNBXnAtf5sVDVt+65SK2C7IxCxZt5o=
+golang.org/x/net v0.56.0/go.mod h1:D3Ku6r+V6JROoZK144D2XfMHFcMq/0zSfLelVTCFKec=
 golang.org/x/oauth2 v0.0.0-20180821212333-d2e6202438be/go.mod h1:N/0e6XlmueqKjAGxoOufVs8QHGRruUQn6yWY3a++T0U=
 golang.org/x/sync v0.0.0-20180314180146-1d60e4601c6f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20181108010431-42b317875d0f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
@@ -369,8 +384,8 @@ golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJ
 golang.org/x/sync v0.0.0-20190911185100-cd5d95a43a6e/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20201020160332-67f06af15bc9/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20210220032951-036812b2e83c/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
-golang.org/x/sync v0.18.0 h1:kr88TuHDroi+UVf+0hZnirlk8o8T+4MrK6mr60WkH/I=
-golang.org/x/sync v0.18.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
+golang.org/x/sync v0.21.0 h1:HLII4xRRTtCRkxYp4HNFF0Js/Og6q2i++KXbg0gHCwM=
+golang.org/x/sync v0.21.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
 golang.org/x/sys v0.0.0-20180823144017-11551d06cbcc/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20180830151530-49385e6e1522/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20180905080454-ebe1bf3edb33/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
@@ -394,16 +409,15 @@ golang.org/x/sys v0.0.0-20210927094055-39ccf1dd6fa6/go.mod h1:oPkhp1MJrh7nUepCBc
 golang.org/x/sys v0.0.0-20220503163025-988cb79eb6c6/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.0.0-20220728004956-3c1f35247d10/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.0.0-20221010170243-090e33056c14/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.38.0 h1:3yZWxaJjBmCWXqhN1qh02AkOnCQ1poK6oF+a7xWL6Gc=
-golang.org/x/sys v0.38.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
+golang.org/x/sys v0.46.0 h1:noSf2Fq6F8DBgS+LysIkx7rIExoNHJsxOAtPp4rthXw=
+golang.org/x/sys v0.46.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
 golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
 golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.2/go.mod h1:bEr9sfX3Q8Zfm5fL9x+3itogRgK3+ptLWKqgva+5dAk=
 golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
-golang.org/x/text v0.31.0 h1:aC8ghyu4JhP8VojJ2lEHBnochRno1sgL6nEi9WGFGMM=
-golang.org/x/text v0.31.0/go.mod h1:tKRAlv61yKIjGGHX/4tP1LTbc13YSec1pxVEWXzfoeM=
+golang.org/x/text v0.38.0 h1:sXmwo9DwP3OK9EZ7PqAdaooSGozfl/3a6/xJcbzPRhE=
+golang.org/x/text v0.38.0/go.mod h1:YXZt3QhHUKYT53r2lLKFIVi6Ao1jdzrTR/KQ09qyxF4=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
 golang.org/x/tools v0.0.0-20190114222345-bf090417da8b/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
 golang.org/x/tools v0.0.0-20190226205152-f727befe758c/go.mod h1:9Yl7xja0Znq3iFh3HoIrodX9oNMXvdceNzlUR8zjMvY=
@@ -413,8 +427,8 @@ golang.org/x/tools v0.0.0-20190907020128-2ca718005c18/go.mod h1:b+2E5dAYhXwXZwtn
 golang.org/x/tools v0.0.0-20191119224855-298f0cb1881e/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
 golang.org/x/tools v0.0.0-20200619180055-7c47624df98f/go.mod h1:EkVYQZoAsY45+roYkvgYkIh4xh/qjgUK9TdY2XT94GE=
 golang.org/x/tools v0.0.0-20210106214847-113979e3529a/go.mod h1:emZCQorbCU4vsT4fOWvOPXz4eW1wZW4PmDk9uLelYpA=
-golang.org/x/tools v0.38.0 h1:Hx2Xv8hISq8Lm16jvBZ2VQf+RLmbd7wVUsALibYI/IQ=
-golang.org/x/tools v0.38.0/go.mod h1:yEsQ/d/YK8cjh0L6rZlY8tgtlKiBNTL14pGDJPJpYQs=
+golang.org/x/tools v0.46.0 h1:7jTurBkPZu4moS/Uy4OQT1M+QBlsj3wejyZwsT8Z7rk=
+golang.org/x/tools v0.46.0/go.mod h1:FrD85F8l+NWL+9XWBSyVSHO6Ne4jutsfIFba7AWQ5Ys=
 golang.org/x/xerrors v0.0.0-20190717185122-a985d3407aa7/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 golang.org/x/xerrors v0.0.0-20191011141410-1b5146add898/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
@@ -429,17 +443,17 @@ google.golang.org/appengine v1.6.7/go.mod h1:8WjMMxjGQR8xUklV/ARdw2HLXBOI7O7uCID
 google.golang.org/genproto v0.0.0-20180817151627-c66870c02cf8/go.mod h1:JiN7NxoALGmiZfu7CAH4rXhgtRTLTxftemlI0sWmxmc=
 google.golang.org/genproto v0.0.0-20190819201941-24fa4b261c55/go.mod h1:DMBHOl98Agz4BDEuKkezgsaosCRResVns1a3J2ZsMNc=
 google.golang.org/genproto v0.0.0-20200526211855-cb27e3aa2013/go.mod h1:NbSheEEYHJ7i3ixzK3sjbqSGDJWnxyFXZblF3eUsNvo=
-google.golang.org/genproto/googleapis/api v0.0.0-20250825161204-c5933d9347a5 h1:BIRfGDEjiHRrk0QKZe3Xv2ieMhtgRGeLcZQ0mIVn4EY=
-google.golang.org/genproto/googleapis/api v0.0.0-20250825161204-c5933d9347a5/go.mod h1:j3QtIyytwqGr1JUDtYXwtMXWPKsEa5LtzIFN1Wn5WvE=
-google.golang.org/genproto/googleapis/rpc v0.0.0-20250825161204-c5933d9347a5 h1:eaY8u2EuxbRv7c3NiGK0/NedzVsCcV6hDuU5qPX5EGE=
-google.golang.org/genproto/googleapis/rpc v0.0.0-20250825161204-c5933d9347a5/go.mod h1:M4/wBTSeyLxupu3W3tJtOgB14jILAS/XWPSSa3TAlJc=
+google.golang.org/genproto/googleapis/api v0.0.0-20260209200024-4cfbd4190f57 h1:JLQynH/LBHfCTSbDWl+py8C+Rg/k1OVH3xfcaiANuF0=
+google.golang.org/genproto/googleapis/api v0.0.0-20260209200024-4cfbd4190f57/go.mod h1:kSJwQxqmFXeo79zOmbrALdflXQeAYcUbgS7PbpMknCY=
+google.golang.org/genproto/googleapis/rpc v0.0.0-20260209200024-4cfbd4190f57 h1:mWPCjDEyshlQYzBpMNHaEof6UX1PmHcaUODUywQ0uac=
+google.golang.org/genproto/googleapis/rpc v0.0.0-20260209200024-4cfbd4190f57/go.mod h1:j9x/tPzZkyxcgEFkiKEEGxfvyumM01BEtsW8xzOahRQ=
 google.golang.org/grpc v1.19.0/go.mod h1:mqu4LbDTu4XGKhr4mRzUsmM4RtVoemTSY81AxZiDr8c=
 google.golang.org/grpc v1.23.0/go.mod h1:Y5yQAOtifL1yxbo5wqy6BxZv8vAUGQwXBOALyacEbxg=
 google.golang.org/grpc v1.25.1/go.mod h1:c3i+UQWmh7LiEpx4sFZnkU36qjEYZ0imhYfXVyQciAY=
 google.golang.org/grpc v1.27.0/go.mod h1:qbnxyOmOxrQa7FizSgH+ReBfzJrCY1pSN7KXBS8abTk=
 google.golang.org/grpc v1.33.2/go.mod h1:JMHMWHQWaTccqQQlmk3MJZS+GWXOdAesneDmEnv2fbc=
-google.golang.org/grpc v1.75.0 h1:+TW+dqTd2Biwe6KKfhE5JpiYIBWq865PhKGSXiivqt4=
-google.golang.org/grpc v1.75.0/go.mod h1:JtPAzKiq4v1xcAB2hydNlWI2RnF85XXcV0mhKXr2ecQ=
+google.golang.org/grpc v1.79.3 h1:sybAEdRIEtvcD68Gx7dmnwjZKlyfuc61Dyo9pGXXkKE=
+google.golang.org/grpc v1.79.3/go.mod h1:KmT0Kjez+0dde/v2j9vzwoAScgEPx/Bw1CYChhHLrHQ=
 google.golang.org/protobuf v0.0.0-20200109180630-ec00e32a8dfd/go.mod h1:DFci5gLYBciE7Vtevhsrf46CRTquxDuWsQurQQe4oz8=
 google.golang.org/protobuf v0.0.0-20200221191635-4d8936d0db64/go.mod h1:kwYJMbMJ01Woi6D6+Kah6886xMZcty6N08ah7+eCXa0=
 google.golang.org/protobuf v0.0.0-20200228230310-ab0ca4ff8a60/go.mod h1:cfTl7dwQJ+fmap5saPgwCLgHXTUD7jkjRqWcaiX5VyM=
@@ -449,8 +463,8 @@ google.golang.org/protobuf v1.22.0/go.mod h1:EGpADcykh3NcUnDUJcl1+ZksZNG86OlYog2
 google.golang.org/protobuf v1.23.0/go.mod h1:EGpADcykh3NcUnDUJcl1+ZksZNG86OlYog2l/sGQquU=
 google.golang.org/protobuf v1.23.1-0.20200526195155-81db48ad09cc/go.mod h1:EGpADcykh3NcUnDUJcl1+ZksZNG86OlYog2l/sGQquU=
 google.golang.org/protobuf v1.25.0/go.mod h1:9JNX74DMeImyA3h4bdi1ymwjUzf21/xIlbajtzgsN7c=
-google.golang.org/protobuf v1.36.8 h1:xHScyCOEuuwZEc6UtSOvPbAT4zRh0xcNRYekJwfqyMc=
-google.golang.org/protobuf v1.36.8/go.mod h1:fuxRtAxBytpl4zzqUh6/eyUujkJdNiuEkXntxiD/uRU=
+google.golang.org/protobuf v1.36.11 h1:fV6ZwhNocDyBLK0dj+fg8ektcVegBBuEolpbTQyBNVE=
+google.golang.org/protobuf v1.36.11/go.mod h1:HTf+CrKn2C3g5S8VImy6tdcUvCska2kB7j23XfzDpco=
 gopkg.in/alecthomas/kingpin.v2 v2.2.6/go.mod h1:FMv+mEhP44yOT+4EoQTLFTRgOQ1FBLkstjWtayDeSgw=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
 gopkg.in/check.v1 v1.0.0-20190902080502-41f04d3bba15/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=

main.go

@@ -26,6 +26,7 @@ func main() {
 
 	// 注册路由
 	http.RouteRegister([]interface{}{
+		controller.ModelManage,
 		controller.ModelGatewayModels,
 		controller.ModelGatewayTask,
 		controller.ModelGatewayLogsStat,
@@ -40,9 +41,10 @@ func main() {
 	<-quit
 
 	g.Log().Infof(ctx, "[main] 收到退出信号，开始优雅退出...")
-	cancel()
-	// 关闭 gateway server（RouteRegister 内部是 go Httpserver.Run() 启动的）
+	// 先关闭 gateway server，等待 in-flight 请求处理完成
 	_ = http.Httpserver.Shutdown()
+	// 再取消上下文，避免活跃请求被中断
+	cancel()
 }
 
 func startAutoRunner(ctx context.Context) {

model/dto/model_gateway_task_dto.go

@@ -13,6 +13,8 @@ type CreateTaskReq struct {
 	RequestPayload map[string]any `p:"requestPayload" json:"requestPayload" dc:"请求负载（透传给模型服务）"`
 	EpicycleId     int64          `json:"epicycleId" dc:"轮次ID"`
 	BuildType      int64          `json:"buildType" dc:"构建类型：1-提示词构建 2-节点构建"`
+	BuildModelName string         `json:"buildModelName" json:"buildModelName" dc:"构建模型名称"`
+	TaskId         string         `json:"taskId" dc:"任务ID"`
 }
 
 type CreateTaskRes struct {

model/dto/model_manage_dto.go

@@ -0,0 +1,127 @@
+package dto
+
+import (
+	"model-gateway/consts/model"
+	"model-gateway/consts/public"
+	"model-gateway/model/entity"
+
+	"gitea.redpowerfuture.com/red-future/common/beans"
+	"github.com/gogf/gf/v2/frame/g"
+)
+
+// CreateModelManageReq 添加模型配置
+type CreateModelManageReq struct {
+	g.Meta                `path:"/createModelManage" method:"post" tags:"new模型管理" summary:"new创建模型配置" dc:"new添加新的模型配置"`
+	ModelSupplier         model.SupplierType       `json:"modelSupplier" v:"required#模型供应商不能为空" dc:"模型供应商"`
+	ModelName             string                   `json:"modelName" v:"required#模型名称不能为空" dc:"模型名称"`
+	ModelType             model.ModelType          `json:"modelType" v:"required#模型类型不能为空" dc:"模型类型"`
+	BaseURL               string                   `json:"baseUrl" v:"required#模型服务地址不能为空" dc:"模型服务地址"`
+	SystemModel           *bool                    `json:"systemModel" dc:"系统模型"`
+	HttpMethod            string                   `json:"httpMethod" dc:"请求方式：GET/POST" d:"POST"`
+	ChatModel             *bool                    `json:"chatModel" dc:"对话模型"`
+	ResponseType          model.ResponseType       `json:"responseType" v:"required#调用模式不能为空" dc:"调用模式：0-同步 1-异步 2-流式"`
+	ApiKey                string                   `json:"apiKey" dc:"调用凭证/密钥"`
+	Enabled               *bool                    `json:"enabled" dc:"启用"`
+	RequestHeadMapping    map[string]string        `json:"requestHeadMapping" dc:"请求头映射"`
+	RequestBodyMapping    map[string]any           `json:"requestBodyMapping" dc:"请求体映射"`
+	ResponseMapping       map[string]any           `json:"responseMapping" dc:"返回映射"`
+	ResponseBodyMapping   map[string]string        `json:"responseBodyMapping" dc:"返回体映射"`
+	MaxConcurrency        int                      `json:"maxConcurrency" dc:"最大并发数（默认10）"`
+	TokenMapping          *entity.TokenMapping     `json:"tokenMapping" dc:"token映射"`
+	AsyncTaskMapping      *entity.AsyncTaskMapping `json:"asyncTaskMapping" dc:"异步任务映射"`
+	TokenPredictPrice     float64                  `json:"tokenPredictPrice" dc:"模型Token预估价格"`
+	TokenPredictPriceUnit string                   `json:"tokenPredictPriceUnit" dc:"模型Token预估价格单位"`
+	MaxTokens             int                      `json:"maxTokens" dc:"最大token数"`
+	MaxDuration           int                      `json:"maxDuration" dc:"最大时长"`
+	LastFrame             string                   `json:"lastFrame" dc:"视频的尾帧图像"`
+}
+
+type CreateModelManageRes struct {
+	Id int64 `json:"id,string" dc:"配置ID"`
+}
+
+type UpdateModelManageReq struct {
+	g.Meta                `path:"/updateModelManage" method:"put" tags:"new模型管理" summary:"new更新模型配置" dc:"new更新指定ID的模型配置"`
+	Id                    int64                    `json:"id" v:"required#id不能为空" dc:"配置ID"`
+	ModelSupplier         model.SupplierType       `json:"modelSupplier" dc:"模型供应商"`
+	ModelName             string                   `json:"modelName" dc:"模型名称"`
+	ModelType             model.ModelType          `json:"modelType" dc:"模型类型"`
+	BaseURL               string                   `json:"baseUrl" dc:"模型服务地址"`
+	SystemModel           *bool                    `json:"systemModel" dc:"系统模型"`
+	HttpMethod            string                   `json:"httpMethod" dc:"请求方式：GET/POST"`
+	ChatModel             *bool                    `json:"chatModel" dc:"对话模型"`
+	ResponseType          model.ResponseType       `json:"responseType" dc:"调用模式：0-同步 1-异步 2-流式"`
+	ApiKey                string                   `json:"apiKey" dc:"调用凭证/密钥"`
+	Enabled               *bool                    `json:"enabled" dc:"启用"`
+	RequestHeadMapping    map[string]string        `json:"requestHeadMapping" dc:"请求头映射"`
+	RequestBodyMapping    map[string]any           `json:"requestBodyMapping" dc:"请求体映射"`
+	ResponseMapping       map[string]any           `json:"responseMapping" dc:"返回映射"`
+	ResponseBodyMapping   map[string]string        `json:"responseBodyMapping" dc:"返回主体映射"`
+	MaxConcurrency        int                      `json:"maxConcurrency" dc:"最大并发数（默认10）"`
+	TokenMapping          *entity.TokenMapping     `json:"tokenMapping" dc:"token映射"`
+	AsyncTaskMapping      *entity.AsyncTaskMapping `json:"asyncTaskMapping" dc:"异步任务映射"`
+	TokenPredictPrice     float64                  `json:"tokenPredictPrice" dc:"模型Token预估价格"`
+	TokenPredictPriceUnit string                   `json:"tokenPredictPriceUnit" dc:"模型Token预估价格单位"`
+	MaxTokens             int                      `json:"maxTokens" dc:"最大token数"`
+	MaxDuration           int                      `json:"maxDuration" dc:"最大时长"`
+	LastFrame             string                   `json:"lastFrame" dc:"视频的尾帧图像"`
+}
+
+type DeleteModelManageReq struct {
+	g.Meta `path:"/deleteModelManage" method:"delete" tags:"new模型管理" summary:"new删除模型配置" dc:"new删除指定ID的模型配置"`
+	Id     int64 `p:"id" json:"id,string" v:"required#id不能为空" dc:"配置ID"`
+}
+
+type GetModelManage struct {
+	ChatModel *bool  `json:"chatModel" dc:"对话模型"`
+	Creator   string `json:"creator" dc:"创建人"`
+	ModelName string `json:"modelName" dc:"模型名称"`
+}
+
+type GetModelManageReq struct {
+	g.Meta `path:"/getModelManage" method:"get" tags:"new模型管理" summary:"new获取模型配置" dc:"new获取指定ID的模型配置"`
+	Id     int64 `p:"id" json:"id,string" v:"required#id不能为空" dc:"配置ID"`
+}
+
+type GetModelManageRes struct {
+	*entity.ModelManage `json:"modelManage"`
+}
+
+// ListModelManageReq 配置列表
+type ListModelManageReq struct {
+	g.Meta      `path:"/listModelManage" method:"get" tags:"new模型管理" summary:"new模型配置列表" dc:"new分页获取模型配置列表"`
+	*beans.Page `json:"page"`
+	ModelName   string `p:"modelName" json:"modelName" dc:"模型名称（模糊查询，可选）"`
+	ModelType   int    `p:"modelType" json:"modelType" dc:"模型类型"`
+	Creator     string `json:"creator" dc:"创建人"`
+}
+
+type ListModelManageRes struct {
+	List  []*entity.ModelManage `json:"list" dc:"列表数据"`
+	Total int                   `json:"total" dc:"总数"`
+}
+
+type CheckChatModelReq struct {
+	g.Meta `path:"/checkChatModel" method:"get" tags:"new模型管理" summary:"new检查是否为聊天模型" dc:"new检查是否为聊天模型"`
+}
+
+type CheckChatModelRes struct {
+	IsChatModel bool `json:"isChatModel" dc:"是否为聊天模型"`
+}
+
+// ModelTypeReq 模型类型列表（分页）
+type ModelTypeReq struct {
+	g.Meta `path:"/modelType" method:"get" tags:"new模型管理" summary:"new模型类型列表" dc:"new分页获取模型类型列表"`
+}
+
+type ModelTypeRes struct {
+	List []*model.TypeTree `json:"list" dc:"模型类型ID到名称的映射"`
+}
+
+type ModelSupplierReq struct {
+	g.Meta `path:"/modelSupplier" method:"get" tags:"new模型管理" summary:"new获取运营商列表" dc:"new获取运营商列表"`
+}
+
+type ModelSupplierRes struct {
+	List []*public.Option `json:"list" dc:"运营商名称到ID的映射"`
+}

model/entity/model_gateway_model.go

@@ -4,99 +4,98 @@ import "gitea.redpowerfuture.com/red-future/common/beans"
 
 type modelGatewayModelCol struct {
 	beans.SQLBaseCol
-	ModelName          string
-	ModelType          string
-	BaseURL            string
-	HttpMethod         string
-	HeadMsg            string
-	FormJSON           string
-	RequestMapping     string
-	ResponseMapping    string
-	ResponseBody       string
-	ResponseTokenField string
-	RequiredFields     string
-	IsPrivate          string
-	IsChatModel        string
-	CallMode           string
-	ApiKey             string
-	Enabled            string
-	MaxConcurrency     string
-	TimeoutSeconds     string
-	RetryTimes         string
-	AutoCleanSeconds   string
-	IsOwner            string
-	OperatorName       string
-	TokenConfig        string
-	ExtendMapping      string
-	QueryConfig        string
-	StreamConfig       string
-	FirstFrame         string
-	LastFrame          string
-	MaxTokens          string
+	ModelName        string
+	ModelType        string
+	BaseURL          string
+	HttpMethod       string
+	HeadMsg          string
+	FormJSON         string
+	RequestMapping   string
+	ResponseMapping  string
+	RequiredFields   string
+	IsPrivate        string
+	IsChatModel      string
+	CallMode         string
+	ApiKey           string
+	Enabled          string
+	MaxConcurrency   string
+	TimeoutSeconds   string
+	RetryTimes       string
+	AutoCleanSeconds string
+	IsOwner          string
+	OperatorName     string
+	TokenConfig      string
+	ExtendMapping    string
+	QueryConfig      string
+	StreamConfig     string
+	FirstFrame       string
+	LastFrame        string
+	BillingConfig    string
 }
 
 var ModelGatewayModelCol = modelGatewayModelCol{
-	SQLBaseCol:         beans.DefSQLBaseCol,
-	ModelName:          "model_name",
-	ModelType:          "model_type",
-	BaseURL:            "base_url",
-	HttpMethod:         "http_method",
-	HeadMsg:            "head_msg",
-	FormJSON:           "form_json",
-	RequestMapping:     "request_mapping",
-	ResponseMapping:    "response_mapping",
-	ResponseBody:       "response_body",
-	ResponseTokenField: "response_token_field",
-	RequiredFields:     "required_fields",
-	IsPrivate:          "is_private",
-	IsChatModel:        "is_chat_model",
-	CallMode:           "call_mode",
-	ApiKey:             "api_key",
-	Enabled:            "enabled",
-	MaxConcurrency:     "max_concurrency",
-	TimeoutSeconds:     "timeout_seconds",
-	RetryTimes:         "retry_times",
-	AutoCleanSeconds:   "auto_clean_seconds",
-	IsOwner:            "is_owner",
-	OperatorName:       "operator_name",
-	TokenConfig:        "token_config",
-	ExtendMapping:      "extend_mapping",
-	QueryConfig:        "query_config",
-	StreamConfig:       "stream_config",
-	FirstFrame:         "first_frame",
-	LastFrame:          "last_frame",
-	MaxTokens:          "max_tokens",
+	SQLBaseCol:       beans.DefSQLBaseCol,
+	ModelName:        "model_name",
+	ModelType:        "model_type",
+	BaseURL:          "base_url",
+	HttpMethod:       "http_method",
+	HeadMsg:          "head_msg",
+	FormJSON:         "form_json",
+	RequestMapping:   "request_mapping",
+	ResponseMapping:  "response_mapping",
+	RequiredFields:   "required_fields",
+	IsPrivate:        "is_private",
+	IsChatModel:      "is_chat_model",
+	CallMode:         "call_mode",
+	ApiKey:           "api_key",
+	Enabled:          "enabled",
+	MaxConcurrency:   "max_concurrency",
+	TimeoutSeconds:   "timeout_seconds",
+	RetryTimes:       "retry_times",
+	AutoCleanSeconds: "auto_clean_seconds",
+	IsOwner:          "is_owner",
+	OperatorName:     "operator_name",
+	TokenConfig:      "token_config",
+	ExtendMapping:    "extend_mapping",
+	QueryConfig:      "query_config",
+	StreamConfig:     "stream_config",
+	FirstFrame:       "first_frame",
+	LastFrame:        "last_frame",
+	BillingConfig:    "billing_config",
 }
 
 type ModelGatewayModel struct {
-	beans.SQLBaseDO    `orm:",inline"`
-	ModelName          string           `orm:"model_name" json:"modelName"`
-	ModelType          int              `orm:"model_type" json:"modelType"`
-	BaseURL            string           `orm:"base_url" json:"baseUrl"`
-	HttpMethod         string           `orm:"http_method" json:"httpMethod"`
-	HeadMsg            map[string]any   `orm:"head_msg" json:"headMsg"`
-	Form               []map[string]any `orm:"form_json" json:"form"`
-	RequestMapping     map[string]any   `orm:"request_mapping" json:"requestMapping"`
-	ResponseMapping    map[string]any   `orm:"response_mapping" json:"responseMapping"`
-	ResponseBody       string           `orm:"response_body" json:"responseBody"`
-	ResponseTokenField string           `orm:"response_token_field" json:"tokenField"`
-	RequiredFields     []string         `orm:"required_fields" json:"requiredFields"`
-	IsPrivate          *int             `orm:"is_private" json:"isPrivate"`
-	IsChatModel        *int             `orm:"is_chat_model" json:"isChatModel"`
-	CallMode           *int             `orm:"call_mode" json:"callMode"`
-	ApiKey             string           `orm:"api_key" json:"apiKey"`
-	Enabled            *int             `orm:"enabled" json:"enabled"`
-	MaxConcurrency     int              `orm:"max_concurrency" json:"maxConcurrency"`
-	TimeoutSeconds     int              `orm:"timeout_seconds" json:"timeoutSeconds"`
-	RetryTimes         int              `orm:"retry_times" json:"retryTimes"`
-	AutoCleanSeconds   int              `orm:"auto_clean_seconds" json:"autoCleanSeconds"`
-	IsOwner            *int             `orm:"is_owner" json:"isOwner"`
-	OperatorName       string           `orm:"operator_name" json:"operatorName"`
-	TokenConfig        map[string]any   `orm:"token_config" json:"tokenConfig"`
-	ExtendMapping      map[string]any   `orm:"extend_mapping" json:"extendMapping"`
-	QueryConfig        map[string]any   `orm:"query_config" json:"queryConfig"`
-	StreamConfig       map[string]any   `orm:"stream_config" json:"streamConfig"`
-	FirstFrame         string           `orm:"first_frame" json:"firstFrame"`
-	LastFrame          string           `orm:"last_frame" json:"lastFrame"`
-	MaxTokens          int              `orm:"max_tokens" json:"maxTokens"`
+	beans.SQLBaseDO  `orm:",inline"`
+	ModelName        string           `orm:"model_name" json:"modelName"`
+	ModelType        int              `orm:"model_type" json:"modelType"`
+	BaseURL          string           `orm:"base_url" json:"baseUrl"`
+	HttpMethod       string           `orm:"http_method" json:"httpMethod"`
+	HeadMsg          map[string]any   `orm:"head_msg" json:"headMsg"`
+	Form             []map[string]any `orm:"form_json" json:"form"`
+	RequestMapping   map[string]any   `orm:"request_mapping" json:"requestMapping"`
+	ResponseMapping  map[string]any   `orm:"response_mapping" json:"responseMapping"`
+	RequiredFields   []string         `orm:"required_fields" json:"requiredFields"`
+	IsPrivate        *int             `orm:"is_private" json:"isPrivate"`
+	IsChatModel      *int             `orm:"is_chat_model" json:"isChatModel"`
+	CallMode         *int             `orm:"call_mode" json:"callMode"`
+	ApiKey           string           `orm:"api_key" json:"apiKey"`
+	Enabled          *int             `orm:"enabled" json:"enabled"`
+	MaxConcurrency   int              `orm:"max_concurrency" json:"maxConcurrency"`
+	TimeoutSeconds   int              `orm:"timeout_seconds" json:"timeoutSeconds"`
+	RetryTimes       int              `orm:"retry_times" json:"retryTimes"`
+	AutoCleanSeconds int              `orm:"auto_clean_seconds" json:"autoCleanSeconds"`
+	IsOwner          *int             `orm:"is_owner" json:"isOwner"`
+	OperatorName     string           `orm:"operator_name" json:"operatorName"`
+	TokenConfig      map[string]any   `orm:"token_config" json:"tokenConfig"`
+	ExtendMapping    map[string]any   `orm:"extend_mapping" json:"extendMapping"`
+	QueryConfig      map[string]any   `orm:"query_config" json:"queryConfig"`
+	StreamConfig     map[string]any   `orm:"stream_config" json:"streamConfig"`
+	FirstFrame       string           `orm:"first_frame" json:"firstFrame"`
+	LastFrame        string           `orm:"last_frame" json:"lastFrame"`
+	BillingConfig    map[string]any   `orm:"billing_config" json:"billingConfig"`
 }
+
+const (
+	ResponseBody = "content"      //返回主体（必填）
+	TotalTokens  = "total_tokens" //总token数
+)

model/entity/model_gateway_task.go

@@ -21,6 +21,8 @@ type modelGatewayTaskCol struct {
 	TmpFile         string
 	RequestPayload  string
 	EpicycleId      string
+	BuildModelName  string
+	BillingData     string
 }
 
 var ModelGatewayTaskCol = modelGatewayTaskCol{
@@ -40,26 +42,30 @@ var ModelGatewayTaskCol = modelGatewayTaskCol{
 	TmpFile:         "tmp_file",
 	RequestPayload:  "request_payload",
 	EpicycleId:      "epicycle_id",
+	BuildModelName:  "build_model_name",
+	BillingData:     "billing_data",
 }
 
 // ModelGatewayTask 模型网关任务
 type ModelGatewayTask struct {
 	beans.SQLBaseDO `orm:",inline"`
-	ModelName       string          `orm:"model_name" json:"modelName"`
-	TaskID          string          `orm:"task_id" json:"taskId"`
-	BizName         string          `orm:"biz_name" json:"bizName"`
-	CallbackURL     string          `orm:"callback_url" json:"callbackUrl"`
-	State           int             `orm:"state" json:"state"`
-	Phase           int             `orm:"phase" json:"phase"`
-	ErrorMsg        string          `orm:"error_msg" json:"errorMsg"`
-	ResultFile      *ResultFile     `orm:"result_file" json:"resultFile"`
-	TextResult      map[string]any  `orm:"text_result" json:"text"`
-	ExpendTokens    int64           `orm:"expend_tokens" json:"expendTokens"`
-	DurationSeconds int64           `orm:"duration_seconds" json:"durationSeconds"`
-	RetryCount      int             `orm:"retry_count" json:"retryCount"`
-	TmpFile         string          `orm:"tmp_file" json:"tmpFile"`
-	RequestPayload  *RequestPayload `orm:"request_payload" json:"requestPayload"`
-	EpicycleId      int64           `orm:"epicycle_id" json:"epicycleId"`
+	ModelName       string           `orm:"model_name" json:"modelName"`
+	TaskID          string           `orm:"task_id" json:"taskId"`
+	BizName         string           `orm:"biz_name" json:"bizName"`
+	CallbackURL     string           `orm:"callback_url" json:"callbackUrl"`
+	State           int              `orm:"state" json:"state"`
+	Phase           int              `orm:"phase" json:"phase"`
+	ErrorMsg        string           `orm:"error_msg" json:"errorMsg"`
+	ResultFile      *ResultFile      `orm:"result_file" json:"resultFile"`
+	TextResult      map[string]any   `orm:"text_result" json:"text"`
+	ExpendTokens    int64            `orm:"expend_tokens" json:"expendTokens"`
+	DurationSeconds int64            `orm:"duration_seconds" json:"durationSeconds"`
+	RetryCount      int              `orm:"retry_count" json:"retryCount"`
+	TmpFile         string           `orm:"tmp_file" json:"tmpFile"`
+	RequestPayload  *RequestPayload  `orm:"request_payload" json:"requestPayload"`
+	EpicycleId      int64            `orm:"epicycle_id" json:"epicycleId"`
+	BuildModelName  string           `orm:"build_model_name" json:"buildModelName"`
+	BillingData     []map[string]any `orm:"billing_data" json:"billingData"`
 }
 
 // ResultFile OSS 结果文件

model/entity/model_manage.go

@@ -0,0 +1,102 @@
+package entity
+
+import (
+	"model-gateway/consts/model"
+
+	"gitea.redpowerfuture.com/red-future/common/beans"
+)
+
+type modelManageCol struct {
+	beans.SQLBaseCol
+	ModelSupplier         string
+	ModelName             string
+	ModelType             string
+	BaseURL               string
+	SystemModel           string
+	HttpMethod            string
+	ChatModel             string
+	ResponseType          string
+	ApiKey                string
+	Enabled               string
+	RequestHeadMapping    string
+	RequestBodyMapping    string
+	ResponseMapping       string
+	ResponseBodyMapping   string
+	MaxConcurrency        string
+	TokenPredictPrice     string
+	TokenPredictPriceUnit string
+	MaxTokens             string
+	MaxDuration           string
+	LastFrame             string
+}
+
+var ModelManageCol = modelManageCol{
+	SQLBaseCol:            beans.DefSQLBaseCol,
+	ModelSupplier:         "model_supplier",
+	ModelName:             "model_name",
+	ModelType:             "model_type",
+	BaseURL:               "base_url",
+	SystemModel:           "system_model",
+	HttpMethod:            "http_method",
+	ChatModel:             "chat_model",
+	ResponseType:          "response_type",
+	ApiKey:                "api_key",
+	Enabled:               "enabled",
+	RequestHeadMapping:    "request_head_mapping",
+	RequestBodyMapping:    "request_body_mapping",
+	ResponseMapping:       "response_mapping",
+	ResponseBodyMapping:   "response_body_mapping",
+	MaxConcurrency:        "max_concurrency",
+	TokenPredictPrice:     "token_predict_price",
+	TokenPredictPriceUnit: "token_predict_price_unit",
+	MaxTokens:             "max_tokens",
+	MaxDuration:           "max_duration",
+	LastFrame:             "last_frame",
+}
+
+type ModelManage struct {
+	beans.SQLBaseDO       `orm:",inline"`
+	ModelSupplier         model.SupplierType `orm:"model_supplier" json:"modelSupplier" description:"模型供应商"`
+	ModelName             string             `orm:"model_name" json:"modelName" description:"模型名称"`
+	ModelType             model.ModelType    `orm:"model_type" json:"modelType" description:"模型类型"`
+	BaseURL               string             `orm:"base_url" json:"baseUrl" description:"模型地址"`
+	SystemModel           *bool              `orm:"system_model" json:"systemModel" description:"系统模型"`
+	HttpMethod            string             `orm:"http_method" json:"httpMethod" description:"http方法"`
+	ChatModel             *bool              `orm:"chat_model" json:"ChatModel" description:"是否聊天模型"`
+	ResponseType          model.ResponseType `orm:"response_type" json:"responseType" description:"返回类型：1同步，2异步，3流"`
+	ApiKey                string             `orm:"api_key" json:"apiKey" description:"api key"`
+	Enabled               *bool              `orm:"enabled" json:"enabled" description:"是否启用"`
+	RequestHeadMapping    map[string]string  `orm:"request_head_mapping" json:"requestHeadMapping" description:"请求头映射"`
+	RequestBodyMapping    map[string]any     `orm:"request_body_mapping" json:"requestBodyMapping" description:"请求体映射"`
+	ResponseMapping       map[string]any     `orm:"response_mapping" json:"responseMapping" description:"响应映射"`
+	ResponseBodyMapping   map[string]string  `orm:"response_body_mapping" json:"responseBodyMapping" description:"响应主体映射"`
+	MaxConcurrency        int                `orm:"max_concurrency" json:"maxConcurrency" description:"最大并发数"`
+	TokenMapping          *TokenMapping      `orm:"token_mapping" json:"tokenMapping" description:"token映射"`
+	AsyncTaskMapping      *AsyncTaskMapping  `orm:"async_task_mapping" json:"asyncTaskMapping" description:"异步任务映射"`
+	TokenPredictPrice     float64            `orm:"token_predict_price" json:"tokenPredictPrice" description:"模型Token预估价格"`
+	TokenPredictPriceUnit string             `orm:"token_predict_price_unit" json:"tokenPredictPriceUnit" description:"模型token预估价格单位（秒，百万Token，千Token，字数）"`
+	MaxTokens             int                `orm:"max_tokens" json:"maxTokens" description:"最大token数"`
+	MaxDuration           int                `orm:"max_duration" json:"maxDuration" description:"最大时长（秒）"`
+	LastFrame             string             `orm:"last_frame" json:"lastFrame" description:"视频的尾帧图像"`
+}
+
+type TokenMapping struct {
+	PromptTokens     string `json:"promptTokens" dc:"输入token"`
+	CompletionTokens string `json:"completionTokens" dc:"输出token"`
+	TotalTokens      string `json:"totalTokens" dc:"总token"`
+}
+
+type AsyncTaskMapping struct {
+	Url                string            `json:"url" dc:"url"`
+	HttpMethod         string            `json:"httpMethod" dc:"http方法" d:"POST"`
+	RequestHeadMapping map[string]string `json:"requestHeadMapping" description:"请求头映射"`
+	ResponseMapping    map[string]any    `json:"responseMapping" description:"响应映射"`
+	TaskId             string            `json:"taskId" dc:"任务id"`
+	TaskStatus         string            `json:"taskStatus" dc:"任务状态"`
+	TaskStatusPending  string            `json:"taskStatusPending" dc:"任务状态-待处理"`
+	TaskStatusRunning  string            `json:"taskStatusRunning" dc:"任务状态-运行中"`
+	TaskStatusSuccess  string            `json:"taskStatusSuccess" dc:"任务状态-成功"`
+	TaskStatusFailed   string            `json:"taskStatusFailed" dc:"任务状态-失败"`
+	TaskStatusCancel   string            `json:"taskStatusCancel" dc:"任务状态-取消"`
+	TaskStatusUnknown  string            `json:"taskStatusUnknown" dc:"任务状态-未知"`
+}

service/gateway/gateway_http_service.go

@@ -7,7 +7,6 @@ import (
 	"errors"
 	"fmt"
 	"mime/multipart"
-	"model-gateway/common/util"
 	"model-gateway/model/entity"
 	"time"
 
@@ -43,16 +42,25 @@ func UploadByTask(ctx context.Context, data []byte, fileExt string) (oss *Upload
 	if err != nil {
 		return nil, err
 	}
-	if _, err := part.Write(data); err != nil {
+	if _, err = part.Write(data); err != nil {
 		return nil, err
 	}
-	contentType := writer.FormDataContentType()
+	//contentType := writer.FormDataContentType()
 	if err = writer.Close(); err != nil {
 		return nil, err
 	}
 
-	headers := util.ForwardHeaders(ctx)
-	headers["Content-Type"] = contentType
+	//headers := util.ForwardHeaders(ctx)
+	//headers["Content-Type"] = contentType
+
+	headers := make(map[string]string)
+	headers["Content-Type"] = writer.FormDataContentType()
+	if r := g.RequestFromCtx(ctx); r != nil {
+		if auth := r.Header.Get("Authorization"); auth != "" {
+			headers["Authorization"] = auth
+		}
+	}
+
 	fullURL := "oss/file/uploadFile"
 	g.Log().Infof(ctx, "[OSS] upload start url=%s filename=%s size=%d", fullURL, filename, len(data))
 
@@ -69,23 +77,35 @@ func UploadByTask(ctx context.Context, data []byte, fileExt string) (oss *Upload
 
 // CallbackPayload 回调请求体
 type CallbackPayload struct {
-	TaskId   string `json:"task_id"`
-	State    int    `json:"state"`
-	OssFile  string `json:"oss_file"`
-	FileType string `json:"file_type"`
-	ErrorMsg string `json:"error_msg"`
+	TaskId      string           `json:"task_id"`
+	State       int              `json:"state"`
+	OssFile     string           `json:"oss_file"`
+	FileType    string           `json:"file_type"`
+	ErrorMsg    string           `json:"error_msg"`
+	BillingDate []map[string]any `json:"billing_data"`
 }
 
 // TriggerCallback 任务的回调
 func TriggerCallback(ctx context.Context, t *entity.ModelGatewayTask) {
-	headers := util.ForwardHeaders(ctx)
+	//headers := util.ForwardHeaders(ctx)
+	headers := make(map[string]string)
+	if r := g.RequestFromCtx(ctx); r != nil {
+		for k, v := range r.Request.Header {
+			if len(v) > 0 {
+				headers[k] = v[0]
+			}
+		}
+	}
 	var resp struct{}
 	payload := CallbackPayload{
-		TaskId:   t.TaskID,
-		State:    t.State,
-		OssFile:  t.ResultFile.OssFile,
-		FileType: t.ResultFile.FileType,
-		ErrorMsg: t.ErrorMsg,
+		TaskId:      t.TaskID,
+		State:       t.State,
+		ErrorMsg:    t.ErrorMsg,
+		BillingDate: t.BillingData,
+	}
+	if !g.IsEmpty(t.ResultFile) {
+		payload.OssFile = t.ResultFile.OssFile
+		payload.FileType = t.ResultFile.FileType
 	}
 	jsonData, err := json.Marshal(payload)
 	if err != nil {
@@ -112,7 +132,15 @@ type PromptsCallbackPayload struct {
 // TriggerPromptsCallback 任务成功后的提示词回调
 func TriggerPromptsCallback(ctx context.Context, t *entity.ModelGatewayTask, epicycleId int64) {
 	callbackURL := "prompts-core/session/callback"
-	headers := util.ForwardHeaders(ctx)
+	//headers := util.ForwardHeaders(ctx)
+	headers := make(map[string]string)
+	if r := g.RequestFromCtx(ctx); r != nil {
+		for k, v := range r.Request.Header {
+			if len(v) > 0 {
+				headers[k] = v[0]
+			}
+		}
+	}
 	var resp struct{}
 	payload := PromptsCallbackPayload{
 		EpicycleId: epicycleId,
@@ -136,7 +164,15 @@ func TriggerPromptsCallback(ctx context.Context, t *entity.ModelGatewayTask, epi
 
 // IsSuperAdmin 调用admin-go服务检查是否是超级管理员
 func IsSuperAdmin(ctx context.Context) (res bool, err error) {
-	headers := util.ForwardHeaders(ctx)
+	//headers := util.ForwardHeaders(ctx)
+	headers := make(map[string]string)
+	if r := g.RequestFromCtx(ctx); r != nil {
+		for k, v := range r.Request.Header {
+			if len(v) > 0 {
+				headers[k] = v[0]
+			}
+		}
+	}
 	var r = make(map[string]bool)
 	if err = commonHttp.Get(ctx, "admin-go/api/v1/system/user/checkIsSuperAdmin", headers, &r); err != nil {
 		return false, err
@@ -144,6 +180,109 @@ func IsSuperAdmin(ctx context.Context) (res bool, err error) {
 	return r["isSuperAdmin"], err
 }
 
+// VideoDurationResp 视频时长接口返回
+type VideoDurationResp struct {
+	Videos           []VideoInfo `json:"videos"`
+	Count            int         `json:"count"`
+	TotalDuration    float64     `json:"totalDuration"`
+	TotalDurationStr string      `json:"totalDurationStr"`
+}
+
+type VideoInfo struct {
+	Index       int     `json:"index"`
+	VideoUrl    string  `json:"videoUrl"`
+	Duration    float64 `json:"duration"`
+	DurationStr string  `json:"durationStr"`
+}
+
+// GetVideoDuration 获取视频时长
+func GetVideoDuration(ctx context.Context, urls []string) (VideoDurationResp, error) {
+	apiURL := "media/video/duration"
+	headers := make(map[string]string)
+	if r := g.RequestFromCtx(ctx); r != nil {
+		for k, v := range r.Request.Header {
+			if len(v) > 0 {
+				headers[k] = v[0]
+			}
+		}
+	}
+
+	body := map[string]any{"video_urls": urls}
+	jsonData, _ := json.Marshal(body)
+
+	var resp VideoDurationResp
+	err := commonHttp.Post(ctx, apiURL, headers, &resp, jsonData)
+	if err != nil {
+		g.Log().Warningf(ctx, "[视频时长] 获取失败 err=%v", err)
+		return resp, err
+	}
+
+	g.Log().Infof(ctx, "[视频时长] 获取成功 count=%d totalDuration=%.2f", resp.Count, resp.TotalDuration)
+	return resp, nil
+}
+
+// DeductBalanceReq 扣减余额请求
+type DeductBalanceReq struct {
+	Id      uint64  `json:"id"`
+	Surplus float64 `json:"surplus"`
+}
+
+// DeductBalance 扣减租户余额
+func DeductBalance(ctx context.Context, tenantId uint64, amount float64) error {
+	apiURL := "admin-go/api/v1/system/tenant/edit"
+	headers := make(map[string]string)
+	if r := g.RequestFromCtx(ctx); r != nil {
+		for k, v := range r.Request.Header {
+			if len(v) > 0 {
+				headers[k] = v[0]
+			}
+		}
+	}
+
+	body := DeductBalanceReq{
+		Id:      tenantId,
+		Surplus: amount,
+	}
+	jsonData, _ := json.Marshal(body)
+
+	var resp struct{}
+	err := commonHttp.Put(ctx, apiURL, headers, &resp, jsonData)
+	if err != nil {
+		g.Log().Warningf(ctx, "[扣减余额] 失败 tenantId=%d amount=%.6f err=%v", tenantId, amount, err)
+		return err
+	}
+	g.Log().Infof(ctx, "[扣减余额] 成功 tenantId=%d amount=%.6f", tenantId, amount)
+	return nil
+}
+
+// TenantSurplusResp 租户余额返回
+type TenantSurplusResp struct {
+	Tenant struct {
+		Surplus float64 `json:"surplus"`
+	} `json:"tenant"`
+}
+
+// GetTenantSurplus 获取租户余额
+func GetTenantSurplus(ctx context.Context, tenantId uint64) (float64, error) {
+	apiURL := fmt.Sprintf("admin-go/api/v1/system/tenant/getTenantDetails?tenantId=%d", tenantId)
+	headers := make(map[string]string)
+	if r := g.RequestFromCtx(ctx); r != nil {
+		for k, v := range r.Request.Header {
+			if len(v) > 0 {
+				headers[k] = v[0]
+			}
+		}
+	}
+
+	var resp TenantSurplusResp
+	err := commonHttp.Get(ctx, apiURL, headers, &resp, nil)
+	if err != nil {
+		g.Log().Warningf(ctx, "[获取余额] 失败 tenantId=%d err=%v", tenantId, err)
+		return 0, err
+	}
+	return resp.Tenant.Surplus, nil
+}
+
 //// callback 向回调地址 POST 任务结果（与查询接口 GetTaskRes 出参一致）
 //func (s *audioTaskService) callback(ctx context.Context, taskID, status, errMsg, callbackURL string) {
 //	if callbackURL == "" {

service/model/model_service.go

@@ -99,7 +99,7 @@ func (s *modelService) Get(ctx context.Context, req *dto.GetModelReq) (*dto.GetM
 		ModelName:   req.ModelName,
 		IsChatModel: req.IsChatModel,
 	})
-	if err != nil {
+	if err != nil || model == nil {
 		return nil, err
 	}
 	return &dto.GetModelRes{

service/model_manage_service.go

@@ -0,0 +1,204 @@
+package service
+
+import (
+	"context"
+	"fmt"
+	"model-gateway/consts/model"
+	"model-gateway/consts/public"
+	"model-gateway/dao"
+	"model-gateway/model/dto"
+	"model-gateway/model/entity"
+	"model-gateway/service/gateway"
+
+	"gitea.redpowerfuture.com/red-future/common/beans"
+	"gitea.redpowerfuture.com/red-future/common/db/gfdb"
+	"gitea.redpowerfuture.com/red-future/common/utils"
+	"github.com/gogf/gf/v2/database/gdb"
+	"github.com/gogf/gf/v2/frame/g"
+	"github.com/gogf/gf/v2/util/gconv"
+)
+
+var ModelManage = &modelManageService{}
+
+type modelManageService struct{}
+
+// Create 创建模型
+func (s *modelManageService) Create(ctx context.Context, req *dto.CreateModelManageReq) (res *dto.CreateModelManageRes, err error) {
+	err = gfdb.DB(ctx, public.DbNameModelGateway).Transaction(ctx, func(ctx context.Context, tx gdb.TX) (err error) {
+		// 1）检查是否是超管
+		var isSuperAdmin bool
+		isSuperAdmin, err = gateway.IsSuperAdmin(ctx)
+		if err != nil {
+			return
+		}
+		req.SystemModel = &isSuperAdmin
+		// 1）如果设为会话模型，先把该用户旧会话模型取消
+		err = s.CancelChatModel(ctx, req.ModelType, req.ChatModel, isSuperAdmin)
+		if err != nil {
+			return
+		}
+		// 2）插入数据
+		id, err := dao.ModelManage.Insert(ctx, req)
+		if err != nil {
+			return
+		}
+		res = &dto.CreateModelManageRes{Id: id}
+		return
+	})
+	return
+}
+
+// Update 更新模型配置
+func (s *modelManageService) Update(ctx context.Context, req *dto.UpdateModelManageReq) (err error) {
+	err = gfdb.DB(ctx, public.DbNameModelGateway).Transaction(ctx, func(ctx context.Context, tx gdb.TX) (err error) {
+		var get *entity.ModelManage
+		get, err = dao.ModelManage.GetNotTenantId(ctx, &dto.GetModelManageReq{
+			Id: req.Id,
+		})
+		if err != nil {
+			return
+		}
+		var user *beans.User
+		user, err = utils.GetUserInfo(ctx)
+		if err != nil {
+			return
+		}
+		// 1）如果不是创建者，且是系统模型，则需要拷贝
+		if get.Creator != user.UserName {
+			if get.SystemModel != nil && *get.SystemModel {
+				if g.IsEmpty(req.ApiKey) {
+					return fmt.Errorf("模型apiKey不能为空")
+				}
+				d := new(dto.CreateModelManageReq)
+				err = gconv.Struct(req, d)
+				if err != nil {
+					return
+				}
+				_, err = s.Create(ctx, d)
+				if err != nil {
+					return err
+				}
+				return
+			}
+			return fmt.Errorf("无权限操作")
+		}
+
+		// 1）检查是否是超管
+		var isSuperAdmin bool
+		isSuperAdmin, err = gateway.IsSuperAdmin(ctx)
+		if err != nil {
+			return
+		}
+		// 1）如果设为会话模型，先把该用户旧会话模型取消
+		err = s.CancelChatModel(ctx, req.ModelType, req.ChatModel, isSuperAdmin)
+		if err != nil {
+			return
+		}
+		// 2）更新数据
+		_, err = dao.ModelManage.Update(ctx, req)
+		return
+	})
+	return
+}
+
+func (s *modelManageService) CancelChatModel(ctx context.Context, modelType model.ModelType, chatModel *bool, isSuperAdmin bool) (err error) {
+	if !g.IsEmpty(chatModel) && *chatModel {
+		if *modelType == *model.ModelTypeInference.Code {
+			if isSuperAdmin {
+				return fmt.Errorf("超级管理员不能设置会话模型")
+			}
+			// 2）获取该用户信息
+			var user *beans.User
+			user, err = utils.GetUserInfo(ctx)
+			if err != nil {
+				return
+			}
+			// 3）取消该用户之前的会话模型
+			var get *entity.ModelManage
+			get, err = dao.ModelManage.Get(ctx, &dto.GetModelManage{
+				Creator:   user.UserName,
+				ChatModel: chatModel,
+			})
+			if err != nil {
+				return
+			}
+			_, err = dao.ModelManage.Update(ctx, &dto.UpdateModelManageReq{
+				Id:        get.Id,
+				ChatModel: gconv.PtrBool(false),
+			})
+			if err != nil {
+				return
+			}
+		} else {
+			return fmt.Errorf("只有推理模型可以设置成会话模型")
+		}
+	}
+	return
+}
+
+// Delete 删除模型
+func (s *modelManageService) Delete(ctx context.Context, req *dto.DeleteModelManageReq) error {
+	_, err := dao.ModelManage.Delete(ctx, req)
+	return err
+}
+
+func (s *modelManageService) Get(ctx context.Context, req *dto.GetModelManageReq) (res *dto.GetModelManageRes, err error) {
+	get, err := dao.ModelManage.GetNotTenantId(ctx, req)
+	if err != nil {
+		return
+	}
+	err = gconv.Struct(get, &res)
+	return
+}
+
+// List 获取模型列表
+func (s *modelManageService) List(ctx context.Context, req *dto.ListModelManageReq) (res *dto.ListModelManageRes, err error) {
+	var user *beans.User
+	user, err = utils.GetUserInfo(ctx)
+	if err != nil {
+		return
+	}
+	req.Creator = user.UserName
+	list, total, err := dao.ModelManage.ListNotTenantId(ctx, req)
+	if err != nil {
+		return
+	}
+	res = &dto.ListModelManageRes{
+		Total: total,
+	}
+	err = gconv.Struct(list, &res.List)
+	return
+}
+
+func (s *modelManageService) CheckChatModel(ctx context.Context, req *dto.CheckChatModelReq) (res *dto.CheckChatModelRes, err error) {
+	user, err := utils.GetUserInfo(ctx)
+	if err != nil {
+		return
+	}
+	get, err := dao.ModelManage.Get(ctx, &dto.GetModelManage{
+		Creator:   user.UserName,
+		ChatModel: gconv.PtrBool(true),
+	})
+	if err != nil {
+		return
+	}
+	res = &dto.CheckChatModelRes{
+		IsChatModel: !g.IsEmpty(get),
+	}
+	return
+}
+
+// GetModelType 获取模型类型
+func (s *modelManageService) GetModelType(ctx context.Context, req *dto.ModelTypeReq) (res *dto.ModelTypeRes, err error) {
+	res = &dto.ModelTypeRes{
+		List: model.GetTypeTreeList(),
+	}
+	return res, nil
+}
+
+// GetModelSupplier 获取运营商列表
+func (s *modelManageService) GetModelSupplier(ctx context.Context, req *dto.ModelSupplierReq) (res *dto.ModelSupplierRes, err error) {
+	return &dto.ModelSupplierRes{
+		List: model.GetSupplierOptionList(),
+	}, nil
+}

service/task/task_service.go

@@ -6,7 +6,6 @@ import (
 	"fmt"
 	"model-gateway/common/util"
 	"model-gateway/consts/public"
-	"model-gateway/service/queue"
 	"time"
 
 	"model-gateway/dao"
@@ -16,7 +15,10 @@ import (
 	"gitea.redpowerfuture.com/red-future/common/beans"
 	"gitea.redpowerfuture.com/red-future/common/utils"
 	"github.com/gogf/gf/v2/database/gdb"
+	"github.com/gogf/gf/v2/database/gredis"
+	"github.com/gogf/gf/v2/errors/gerror"
 	"github.com/gogf/gf/v2/frame/g"
+	"github.com/gogf/gf/v2/os/glog"
 	"github.com/gogf/gf/v2/util/gconv"
 	"github.com/google/uuid"
 )
@@ -27,13 +29,19 @@ type taskService struct{}
 
 // Create 创建任务
 func (s *taskService) Create(ctx context.Context, req *dto.CreateTaskReq) (res *dto.CreateTaskRes, err error) {
-	taskID := uuid.NewString()
+	taskID := req.TaskId
+	if taskID == "" {
+		taskID = uuid.NewString()
+	}
+	startAt := time.Now()
 
-	// 1) 检查模型配置，并且获取模型
+	// 1) 获取用户信息
 	userInfo, err := utils.GetUserInfo(ctx)
 	if err != nil {
 		return nil, err
 	}
+
+	// 2) 检查模型配置
 	model, err := dao.ModelGatewayModels.Get(ctx, &entity.ModelGatewayModel{
 		SQLBaseDO: beans.SQLBaseDO{
 			TenantId: userInfo.TenantId,
@@ -47,78 +55,196 @@ func (s *taskService) Create(ctx context.Context, req *dto.CreateTaskReq) (res *
 	if model == nil || (model.Enabled != nil && *model.Enabled != 1) {
 		return nil, errors.New("模型不存在或未启用")
 	}
+	lockKey := fmt.Sprintf("lock:tenantId-%s:model-%s", gconv.String(userInfo.TenantId), req.ModelName)
+	success, e := Lock(ctx, lockKey, -1, int64(time.Minute.Seconds()*5), func(ctx context.Context) error {
+		const (
+			keyExpireSec = 600              // 计数Key兜底过期时间 10min
+			waitInterval = 10 * time.Second // 轮询等待间隔
+		)
+		// Redis 操作统一使用独立上下文，避免外部 ctx canceled
+		redisCtx := context.WithoutCancel(ctx)
+		// 模型并发计数Key
+		concurrencyKey := fmt.Sprintf("model:concurrency:%s", req.ModelName)
+		maxCon := gconv.Int64(model.MaxConcurrency)
 
-	// 2) 排队上限（严格控制：Redis 原子闸门）
-	limit := queue.GetRuntimeQueueLimit(ctx, req.ModelName, model.MaxConcurrency*2)
-	if limit > 0 {
-		ok, err := queue.AcquireQueueSlot(ctx, req.ModelName, taskID, limit, model.TimeoutSeconds)
-		if err != nil {
-			return nil, err
+		// 循环尝试获取并发名额，超限则等待重试
+		var held bool // 标记当前是否持有未释放的计数
+		for {
+			// 检测全局上下文取消
+			if ctx.Err() != nil {
+				if held {
+					g.Redis().Decr(redisCtx, concurrencyKey)
+				}
+				return ctx.Err()
+			}
+			// 计数自增
+			currentCon, e := g.Redis().Incr(redisCtx, concurrencyKey)
+			if e != nil {
+				if held {
+					g.Redis().Decr(redisCtx, concurrencyKey)
+				}
+				glog.Errorf(ctx, "redis incr concurrency key err: %v", e)
+				return e
+			}
+			held = true
+			// 首次创建Key时设置过期时间（避免重复执行EXPIRE）
+			exists, errr := g.Redis().Exists(redisCtx, concurrencyKey)
+			if errr == nil && exists == 1 {
+				g.Redis().Expire(redisCtx, concurrencyKey, keyExpireSec)
+			}
+
+			// 未超限：跳出循环，执行业务
+			if currentCon <= maxCon {
+				glog.Infof(ctx, "并发数: %s %d/%d", concurrencyKey, currentCon, maxCon)
+				break
+			}
+			// 超限立刻回减，撤销本次计数
+			g.Redis().Decr(redisCtx, concurrencyKey)
+			held = false
+			glog.Infof(ctx, "并发超限等待: %s %d/%d", concurrencyKey, currentCon, maxCon)
+			time.Sleep(waitInterval)
 		}
-		if !ok {
-			return nil, errors.New("任务排队已满，请稍后再试")
+
+		// 3) 构建任务实体
+		task := &entity.ModelGatewayTask{
+			ModelName:   model.ModelName,
+			TaskID:      taskID,
+			State:       public.TaskStatusRunning,
+			BizName:     req.BizName,
+			CallbackURL: req.CallbackUrl,
+			RequestPayload: &entity.RequestPayload{
+				Body:    req.RequestPayload,
+				Headers: util.ParseHeadMsgHeaders(model.HeadMsg),
+			},
+			EpicycleId:     req.EpicycleId,
+			BuildModelName: req.BuildModelName,
 		}
-	}
 
-	// 3) 插入任务记录
-	requestPayload := entity.RequestPayload{
-		Body:    req.RequestPayload,
-		Headers: util.ParseHeadMsgHeaders(model.HeadMsg),
-	}
-	id, err := dao.ModelGatewayTask.Insert(ctx, &entity.ModelGatewayTask{
-		ModelName:      req.ModelName,
-		TaskID:         taskID,
-		State:          public.TaskStatusPending,
-		BizName:        req.BizName,
-		CallbackURL:    req.CallbackUrl,
-		RequestPayload: &requestPayload,
-		EpicycleId:     req.EpicycleId,
+		// 4) 插入任务记录
+		id, errr := dao.ModelGatewayTask.Insert(ctx, task)
+		if errr != nil {
+			g.Redis().Decr(redisCtx, concurrencyKey)
+			// TODO: 恢复排队逻辑后，此处需要回滚排队占位
+			//queue.ReleaseQueueSlot(ctx, req.ModelName, taskID)
+			return errr
+		}
+		task.Id = id
+
+		// 5) 记录操作日志（非关键路径，失败不影响主流程）
+		ip, ua := "", ""
+		if r := g.RequestFromCtx(ctx); r != nil {
+			ip = utils.GetLocalIP()
+			ua = r.UserAgent()
+		}
+		_, _ = dao.ModelGatewayLogsOp.Insert(ctx, &entity.ModelGatewayLogsOp{
+			IP:              ip,
+			UserAgent:       ua,
+			APIPath:         "/task/createTask",
+			HttpMethod:      "POST",
+			BizName:         req.BizName,
+			ModelName:       req.ModelName,
+			TaskID:          taskID,
+			OpType:          "createTask",
+			Success:         1,
+			CostMs:          time.Since(startAt).Milliseconds(),
+			RequestPayload:  task.RequestPayload,
+			ResponsePayload: gdb.Map{"taskId": taskID},
+		})
+
+		// 6) 模型计费
+		if len(model.BillingConfig) > 0 {
+			requestData := util.ExtractRequestBilling(ctx, model.BillingConfig, req.RequestPayload)
+			// 请求数据作为计费记录的基础字段，先存入数组
+			task.BillingData = append(task.BillingData, requestData)
+			_, _ = dao.ModelGatewayTask.Update(ctx, &entity.ModelGatewayTask{
+				SQLBaseDO:   beans.SQLBaseDO{Id: task.Id},
+				BillingData: task.BillingData,
+			})
+		}
+
+		// 7) 异步执行任务
+		go AsyncWorker.handleOne(util.AsyncCtx(ctx), task, model, req)
+
+		return nil
 	})
-	if err != nil { // 入库失败：回滚闸门占位
-		queue.ReleaseQueueSlot(ctx, req.ModelName, taskID)
-		return nil, err
+	if e != nil {
+		err = e
+		return
 	}
-
-	// 4) 写操作日志（不影响主流程，失败忽略）
-	ip := ""
-	ua := ""
-	apiPath := "/task/createTask"
-	httpMethod := "POST"
-	if r := g.RequestFromCtx(ctx); r != nil {
-		ip = utils.GetLocalIP()
-		ua = r.UserAgent()
-		apiPath = r.URL.Path
-		httpMethod = r.Method
+	if !success {
+		err = gerror.New("任务排队已满，请稍后再试")
+		return
 	}
-	_, _ = dao.ModelGatewayLogsOp.Insert(ctx, &entity.ModelGatewayLogsOp{
-		IP:             ip,
-		UserAgent:      ua,
-		APIPath:        apiPath,
-		HttpMethod:     httpMethod,
-		BizName:        req.BizName,
-		ModelName:      req.ModelName,
-		TaskID:         taskID,
-		OpType:         "createTask",
-		Success:        1,
-		CostMs:         time.Since(time.Now()).Milliseconds(),
-		RequestPayload: &requestPayload,
-		ResponsePayload: gdb.Map{
-			"taskId": taskID,
-		},
-	})
-
-	// 5) 获取任务信息
-	task, err := dao.ModelGatewayTask.ClaimByID(ctx, id)
-	if err != nil {
-		return nil, err
-	}
-
-	// 5) 创建成功后立即异步尝试执行当前任务
-	go AsyncWorker.handleOne(util.AsyncCtx(ctx), task, model, req)
 
 	return &dto.CreateTaskRes{TaskID: taskID}, nil
 }
 
+// Lock 分布式锁 纯原生命令、无Lua、隔离上下文防 context canceled
+func Lock(ctx context.Context, key string, limit, expireSeconds int64, fn func(ctx context.Context) error) (success bool, err error) {
+	if limit <= 0 {
+		limit = -1
+	}
+
+	// 过期时间合法校验（单位：秒）
+	const maxExpireSec = 86400 * 7
+	if expireSeconds < 1 || expireSeconds > maxExpireSec {
+		glog.Warningf(ctx, "锁过期时间非法，原值:%d，兜底为60秒", expireSeconds)
+		expireSeconds = 60
+	}
+
+	lockVal := "1"
+
+LOOP:
+	// 检测父级上下文取消，防止无限重试阻塞 goroutine
+	if ctx.Err() != nil {
+		return false, ctx.Err()
+	}
+	if limit != -1 {
+		if limit < 0 {
+			return false, errors.New("锁重试次数耗尽，获取锁失败")
+		}
+		limit--
+	}
+
+	// 核心：创建独立上下文，不受外部 ctx 取消影响
+	redisCtx := context.WithoutCancel(ctx)
+
+	// 加锁
+	val, err := g.Redis().Set(redisCtx, key, lockVal, gredis.SetOption{
+		TTLOption: gredis.TTLOption{
+			EX: &expireSeconds,
+		},
+		NX: true,
+	})
+	if err != nil {
+		glog.Errorf(ctx, "redis set lock failed: %v", err)
+		time.Sleep(time.Second)
+		goto LOOP
+	}
+
+	if val.Bool() {
+		// 执行业务逻辑（使用原上下文）
+		runErr := fn(ctx)
+
+		// 释放锁：同样使用独立上下文 + 先GET再DEL防误删
+		getRes, err := g.Redis().Get(redisCtx, key)
+		if err != nil {
+			glog.Errorf(ctx, "redis get lock value failed: %v", err)
+		} else if getRes.String() == lockVal {
+			_, delErr := g.Redis().Del(redisCtx, key)
+			if delErr != nil {
+				glog.Errorf(ctx, "redis del lock failed: %v", delErr)
+			}
+		}
+
+		return true, runErr
+	}
+
+	// 抢锁失败，休眠重试
+	time.Sleep(time.Second)
+	goto LOOP
+}
+
 // GetResult 获取任务结果
 func (s *taskService) GetResult(ctx context.Context, taskID string) (res *dto.GetTaskResultRes, err error) {
 	t, err := dao.ModelGatewayTask.Get(ctx, &entity.ModelGatewayTask{
@@ -255,7 +381,10 @@ func (s *taskService) QueryPendingTasks(ctx context.Context, req *dto.QueryPendi
 		if err != nil || model == nil || model.QueryConfig == nil {
 			continue
 		}
-		result, err := util.PullTaskResult(ctx, nil, model.QueryConfig, model.HeadMsg)
+		// 每个任务使用独立的超时上下文，防止单个任务阻塞整个轮询
+		pullCtx, pullCancel := context.WithTimeout(ctx, 30*time.Second)
+		result, err := util.PullTaskResult(pullCtx, nil, model.QueryConfig, model.HeadMsg)
+		pullCancel()
 		if err != nil {
 			g.Log().Warningf(ctx, "[轮询] 查询失败 taskID=%s err=%v", t.TaskID, err)
 			continue

service/task/worker.go

@@ -6,22 +6,19 @@ import (
 	"encoding/json"
 	"fmt"
 	"io"
-	"net/http"
-	"os"
-	"strings"
-	"sync"
-	"time"
-	"unicode/utf8"
-
 	"model-gateway/common/util"
 	"model-gateway/consts/public"
 	"model-gateway/dao"
 	"model-gateway/model/dto"
 	"model-gateway/model/entity"
 	"model-gateway/service/gateway"
-	"model-gateway/service/queue"
+	"net/http"
+	"strings"
+	"sync"
+	"time"
 
 	"gitea.redpowerfuture.com/red-future/common/beans"
+	"gitea.redpowerfuture.com/red-future/common/utils"
 	"github.com/gogf/gf/v2/encoding/gjson"
 	"github.com/gogf/gf/v2/frame/g"
 	"github.com/gogf/gf/v2/util/gconv"
@@ -38,73 +35,125 @@ func (w *asyncWorker) handleOne(ctx context.Context, task *entity.ModelGatewayTa
 		body      = task.RequestPayload.Body
 		maxRetry  = model.RetryTimes
 		startTime = time.Now()
+		rawData   []byte
 		result    map[string]any
 		err       error
+		surplus   float64
 	)
-	g.Log().Infof(ctx, "[执行任务][开始] taskId=%s model=%s", task.TaskID, task.ModelName)
+	g.Log().Infof(ctx, "[handleOne] 开始 taskId=%s model=%s", task.TaskID, task.ModelName)
 
 	// ============================================
-	// 1) 分布式并发控制
+	// 1) 查询余额
 	// ============================================
-	semKey := fmt.Sprintf("asynch:sem:%s", task.ModelName)
-	maxC := queue.GetRuntimeMaxConcurrency(ctx, task.ModelName, model.MaxConcurrency)
-	acquired, err := queue.AcquireSemaphore(ctx, semKey, maxC, 3600)
-	if err != nil {
-		task.DurationSeconds = int64(time.Since(startTime).Seconds())
-		w.failTask(ctx, task, startTime, err.Error())
+	surplus, _ = gateway.GetTenantSurplus(ctx, model.TenantId)
+	if surplus <= 200 {
+		w.failTask(ctx, task, startTime, "租户余额不足")
 		return
 	}
-	if !acquired {
-		_, _ = dao.ModelGatewayTask.Update(ctx, &entity.ModelGatewayTask{
-			SQLBaseDO: beans.SQLBaseDO{Id: task.Id},
-			State:     public.TaskStatusPending,
-		})
-		g.Log().Infof(ctx, "[执行任务][排队] 并发已满，放回队列 taskId=%s", task.TaskID)
-		return
-	}
-	defer func() { _ = queue.ReleaseSemaphore(ctx, semKey) }()
+	g.Log().Infof(ctx, "[handleOne] 当前余额 tenantId=%d surplus=%.2f", model.TenantId, surplus)
 
 	// ============================================
 	// 2) 调用模型
 	// ============================================
-	switch {
-	case model.CallMode != nil && *model.CallMode == public.CallModeStream:
-		rawBytes, streamErr := w.callModelStream(ctx, task, model, body)
-		if streamErr != nil {
-			w.failTask(ctx, task, startTime, streamErr.Error())
+	for attempt := 0; ; attempt++ {
+		if attempt > 0 {
+			g.Log().Infof(ctx, "[handleOne] 调模型重试 第%d次 taskId=%s", attempt, task.TaskID)
+			time.Sleep(time.Duration(attempt) * 5 * time.Second)
+		}
+
+		rawData, err = InvokeModel(ctx, model, body)
+		switch {
+		case model.CallMode != nil && *model.CallMode == public.CallModeStream:
+			if err == nil {
+				result, err = util.ParseStreamResponse(rawData, model.StreamConfig)
+			}
+		case model.CallMode != nil && *model.CallMode == public.CallModeAsync:
+			if err == nil {
+				result = gjson.New(string(rawData)).Map()
+				result, err = util.PullTaskResult(ctx, result, model.QueryConfig, model.HeadMsg)
+			}
+		default:
+			if err == nil {
+				result = gjson.New(string(rawData)).Map()
+			}
+		}
+
+		if err == nil {
+			break
+		}
+		// 模型调用失败
+		if !strings.Contains(err.Error(), "Timeout") &&
+			!strings.Contains(err.Error(), "RequestCanceled") &&
+			!strings.Contains(err.Error(), "InternalServiceError") &&
+			!strings.Contains(err.Error(), "Invalid video_url") &&
+			!strings.Contains(err.Error(), "Invalid audio track") &&
+			!strings.Contains(err.Error(), "Error while downloading") &&
+			!strings.Contains(err.Error(), "Error while connecting") &&
+			!strings.Contains(err.Error(), "download failed") {
+			w.failTask(ctx, task, startTime, err.Error())
 			return
 		}
-		result, err = util.ParseStreamResponse(rawBytes, model.StreamConfig)
-	case model.CallMode != nil && *model.CallMode == public.CallModeAsync:
-		result, err = w.callModel(ctx, task, model, body)
-		if err == nil {
-			result, err = util.PullTaskResult(ctx, result, model.QueryConfig, model.HeadMsg)
+
+		g.Log().Warningf(ctx, "[handleOne] 调模型失败 taskId=%s attempt=%d err=%v", task.TaskID, attempt, err)
+	}
+
+	// ============================================
+	// 3) 解析返回映射 + 存储 token 相关信息
+	// ============================================
+	mapped, err := util.MapResponsePayload(model.ResponseMapping, result)
+	if err != nil {
+		w.failTask(ctx, task, startTime, err.Error())
+		return
+	}
+
+	// 计费处理
+	if len(model.BillingConfig) > 0 && len(task.BillingData) > 0 {
+		// 取请求阶段数据作为基础
+		billingInput := make(map[string]any)
+		for k, v := range task.BillingData[0] {
+			billingInput[k] = v
+		}
+		// 补充返回数据
+		responseData := util.ExtractResponseBilling(model.BillingConfig, mapped)
+		for k, v := range responseData {
+			billingInput[k] = v
+		}
+		// 计算费用，替换数组第一个元素
+		billingResult := util.CalculateBilling(model.BillingConfig, billingInput)
+		if billingResult != nil {
+			task.BillingData[0] = billingResult
+		}
+		if billingResult != nil {
+			task.BillingData[0] = billingResult
+			totalFee := gconv.Float64(billingResult["total_fee"])
+			if totalFee > 0 {
+				for attempt := 0; attempt <= maxRetry; attempt++ {
+					err = gateway.DeductBalance(util.AsyncCtx(ctx), model.TenantId, -totalFee)
+					if err == nil {
+						break
+					}
+					g.Log().Warningf(ctx, "[handleOne] 扣除余额失败 taskId=%s attempt=%d/%d err=%v", task.TaskID, attempt, maxRetry, err)
+				}
+			}
 		}
-	default:
-		result, err = w.callModel(ctx, task, model, body)
 	}
-	if err != nil {
-		w.failTask(ctx, task, startTime, err.Error())
+
+	task.ExpendTokens = gconv.Int64(mapped[entity.TotalTokens])
+	if _, err = dao.ModelGatewayTask.Update(ctx, task); err != nil {
+		g.Log().Errorf(ctx, "[handleOne] 更新DB失败 taskId=%s err=%v", task.TaskID, err)
 		return
 	}
 
 	// ============================================
-	// 3) 缓存临时文件
+	// 4) 处理提示词相关数据解析涵盖重试
 	// ============================================
-	if tmpPath, tmpErr := util.SaveTempFileByType(task.TaskID, result, task.TmpFile); tmpErr == nil && tmpPath != "" {
-		task.TmpFile = tmpPath
-		task.Phase = 1
-		_, _ = dao.ModelGatewayTask.Update(ctx, task)
-	}
-
-	// ============================================
-	// 4) 解析校验 + 响应映射（可重试）
-	// ============================================
-	result, err = w.parseAndRetry(ctx, result, task, model, req, maxRetry, startTime)
-	if err != nil {
-		task.TextResult = result
-		w.failTask(ctx, task, startTime, err.Error())
-		return
+	if req.BuildType == public.BuildTypePrompt {
+		mapped, err = w.parseAndRetry(ctx, mapped, model, task, maxRetry)
+		if err != nil {
+			task.TextResult = mapped
+			w.failTask(ctx, task, startTime, err.Error())
+			return
+		}
 	}
 
 	// ============================================
@@ -113,18 +162,14 @@ func (w *asyncWorker) handleOne(ctx context.Context, task *entity.ModelGatewayTa
 	var oss *gateway.UploadFileResponse
 	for attempt := 0; attempt <= maxRetry; attempt++ {
 		if attempt > 0 {
-			g.Log().Infof(ctx, "[执行任务][重试] OSS上传 第%d/%d次 taskId=%s", attempt, maxRetry, task.TaskID)
+			g.Log().Infof(ctx, "[handleOne] OSS上传重试 第%d/%d次 taskId=%s", attempt, maxRetry, task.TaskID)
 		}
-		oss, err = gateway.UploadByTask(ctx, gjson.New(result).MustToJson(), "json")
+		oss, err = gateway.UploadByTask(ctx, gjson.New(mapped).MustToJson(), "json")
 		if err == nil {
 			break
 		}
-		g.Log().Errorf(ctx, "[执行任务][失败] OSS上传失败 taskId=%s attempt=%d/%d err=%v", task.TaskID, attempt, maxRetry, err)
+		g.Log().Errorf(ctx, "[handleOne] OSS上传失败 taskId=%s attempt=%d/%d err=%v", task.TaskID, attempt, maxRetry, err)
 		if attempt == maxRetry {
-			task.State = public.TaskStatusFailed
-			task.ErrorMsg = err.Error()
-			task.Phase = 1
-			_, _ = dao.ModelGatewayTask.Update(ctx, task)
 			w.failTask(ctx, task, startTime, fmt.Sprintf("OSS上传重试耗尽: %v", err))
 			return
 		}
@@ -140,53 +185,20 @@ func (w *asyncWorker) handleOne(ctx context.Context, task *entity.ModelGatewayTa
 		FileType: oss.FileFormat,
 		FileSize: int64(oss.FileSize),
 	}
-	task.TextResult = result
+	task.TextResult = mapped
 	if _, err = dao.ModelGatewayTask.Update(ctx, task); err != nil {
-		g.Log().Errorf(ctx, "[执行任务][失败] 更新数据库失败 taskId=%s err=%v", task.TaskID, err)
+		g.Log().Errorf(ctx, "[handleOne] 更新DB失败 taskId=%s err=%v", task.TaskID, err)
 		return
 	}
-
-	queue.ReleaseQueueSlot(ctx, task.ModelName, task.TaskID)
-	go gateway.TriggerCallback(context.WithoutCancel(ctx), task)
+	concurrencyKey := fmt.Sprintf("model:concurrency:%s", req.ModelName)
+	g.Redis().Decr(ctx, concurrencyKey)
+	gateway.TriggerCallback(ctx, task)
 	if req.EpicycleId != 0 {
-		go gateway.TriggerPromptsCallback(context.WithoutCancel(ctx), task, req.EpicycleId)
+		gateway.TriggerPromptsCallback(ctx, task, req.EpicycleId)
 	}
 
-	g.Log().Infof(ctx, "[执行任务][成功] taskId=%s duration=%ds fileType=%s",
+	g.Log().Infof(ctx, "[handleOne] 成功 taskId=%s duration=%ds fileType=%s",
 		task.TaskID, task.DurationSeconds, oss.FileFormat)
-
-	_ = os.Remove(task.TmpFile)
-}
-
-// callModelStream 调用模型，返回原始字节（不做响应映射，用于流式输出）
-func (w *asyncWorker) callModelStream(ctx context.Context, task *entity.ModelGatewayTask, model *entity.ModelGatewayModel, body map[string]any) ([]byte, error) {
-	var data []byte
-	var err error
-
-	if task.Phase == 1 && strings.TrimSpace(task.TmpFile) != "" {
-		data, err = os.ReadFile(task.TmpFile)
-		if err != nil || len(data) == 0 {
-			data = nil
-		}
-	}
-
-	if data == nil {
-		data, err = InvokeModel(ctx, model, body)
-		if err != nil {
-			return nil, err
-		}
-		tmpPath, tmpErr := util.SaveTmpResult(task.TaskID, data, "")
-		if tmpErr == nil && tmpPath != "" {
-			task.TmpFile = tmpPath
-			task.Phase = 1
-			_, err = dao.ModelGatewayTask.Update(ctx, task)
-			if err != nil {
-				g.Log().Errorf(ctx, "[执行任务][失败] 临时文件保存失败 taskId=%s err=%v", task.TaskID, tmpErr)
-			}
-		}
-	}
-
-	return data, nil
 }
 
 // asyncResult 异步任务结果
@@ -200,12 +212,13 @@ var asyncTaskChan = sync.Map{} // taskID → chan asyncResult
 
 func (w *asyncWorker) callModelAsync(ctx context.Context, task *entity.ModelGatewayTask, model *entity.ModelGatewayModel, body map[string]any) (map[string]any, error) {
 	// 1. 提交异步任务
-	body, err := w.callModel(ctx, task, model, body)
+	rawData, err := InvokeModel(ctx, model, body)
 	if err != nil {
 		return nil, err
 	}
+	body = gjson.New(string(rawData)).Map()
 	// 2. 拿到 task_id
-	taskID := gjson.New(body).Get(model.ResponseBody).String()
+	taskID := gjson.New(body).Get(entity.ResponseBody).String()
 
 	// 3. 创建等待通道
 	ch := make(chan asyncResult, 1)
@@ -241,135 +254,170 @@ func NotifyAsyncResult(taskID string, result map[string]any, err error) {
 	}
 }
 
-// callModel 调用模型 + 检测文件类型 + 保存临时文件
-// 返回: 解析后的响应体, error
-func (w *asyncWorker) callModel(ctx context.Context, task *entity.ModelGatewayTask, model *entity.ModelGatewayModel, body map[string]any) (map[string]any, error) {
-	var data []byte
-	var err error
-
-	// 1) 如果已有临时文件且 phase=1，直接读取
-	if task.Phase == 1 && strings.TrimSpace(task.TmpFile) != "" {
-		data, err = os.ReadFile(task.TmpFile)
-		if err != nil || len(data) == 0 {
-			g.Log().Warningf(ctx, "[callModel] 读取临时文件失败，重新调用模型 taskId=%s err=%v", task.TaskID, err)
-			data = nil
-		}
-	}
-
-	// 2) 没有可用数据，调用模型
-	if data == nil {
-		data, err = InvokeModel(ctx, model, body)
-		if err != nil {
-			return nil, err
-		}
-
-		// 3) 检测文件类型，保存临时文件
-		_, ext := util.DetectFileType(data)
-		tmpPath, tmpErr := util.SaveTmpResult(task.TaskID, data, ext)
-		if tmpErr == nil && tmpPath != "" {
-			task.TmpFile = tmpPath
-			task.Phase = 1
-			_, err = dao.ModelGatewayTask.Update(ctx, task)
-			if err != nil {
-				g.Log().Errorf(ctx, "[执行任务][失败] 临时文件保存失败 taskId=%s err=%v", task.TaskID, tmpErr)
-			}
-		}
-	}
-
-	// 4) 检测文件类型，提取文本结果
-	contentType, _ := util.DetectFileType(data)
-	var textResult string
-	if utf8.Valid(data) && (strings.HasPrefix(contentType, "text/") || contentType == "application/json") {
-		textResult = string(data)
-	}
-
-	// 5) 非文本内容，返回错误
-	if textResult == "" {
-		return nil, fmt.Errorf("模型返回非文本内容，contentType=%s", contentType)
-	}
-
-	// 6) 解析并返回
-	return gjson.New(textResult).Map(), nil
-}
+//// callModel 调用模型 + 提取文本结果
+//func (w *asyncWorker) callModel(ctx context.Context, task *entity.ModelGatewayTask, model *entity.ModelGatewayModel, body map[string]any) (map[string]any, error) {
+//	data, err := InvokeModel(ctx, model, body)
+//	if err != nil {
+//		return nil, err
+//	}
+//	contentType, _ := util.DetectFileType(data)
+//	var textResult string
+//	if utf8.Valid(data) && (strings.HasPrefix(contentType, "text/") || contentType == "application/json") {
+//		textResult = string(data)
+//	}
+//
+//	if textResult == "" {
+//		return nil, fmt.Errorf("模型返回非文本内容，contentType=%s", contentType)
+//	}
+//
+//	return gjson.New(textResult).Map(), nil
+//}
 
 // parseAndRetry 解析模型返回结果，并重试
-func (w *asyncWorker) parseAndRetry(ctx context.Context, body map[string]any, task *entity.ModelGatewayTask, model *entity.ModelGatewayModel, req *dto.CreateTaskReq, maxRetry int, startTime time.Time) (map[string]any, error) {
+func (w *asyncWorker) parseAndRetry(ctx context.Context, body map[string]any, model *entity.ModelGatewayModel, task *entity.ModelGatewayTask, maxRetry int) (map[string]any, error) {
+	// 获取构建模型的必填字段
+	user, err := utils.GetUserInfo(ctx)
+	if err != nil {
+		return nil, err
+	}
+	buildModel, err := dao.ModelGatewayModels.Get(ctx, &entity.ModelGatewayModel{
+		SQLBaseDO: beans.SQLBaseDO{TenantId: user.TenantId, Creator: user.UserName},
+		ModelName: task.BuildModelName,
+	})
+	if err != nil {
+		return nil, err
+	}
+
+	var lastErr error
 	for attempt := 0; attempt <= maxRetry; attempt++ {
 		if attempt > 0 {
 			g.Log().Infof(ctx, "[执行任务][重试] JSON解析 第%d/%d次 taskId=%s", attempt, maxRetry, task.TaskID)
 		}
 
-		// 1) 响应映射
-		mapped, err := util.MapResponsePayload(model.ResponseMapping, body)
-		if err != nil {
-			g.Log().Warningf(ctx, "[执行任务][映射失败] taskId=%s attempt=%d/%d err=%v", task.TaskID, attempt, maxRetry, err)
-			if attempt == maxRetry {
-				return nil, fmt.Errorf("响应映射重试耗尽: %w", err)
-			}
-			continue
-		}
-
-		// 2) 先存 token 到数据库，防止后续失败丢失
-		if _, ok := mapped[model.ResponseTokenField]; ok {
-			task.ExpendTokens = gconv.Int64(mapped[model.ResponseTokenField])
-			_, err = dao.ModelGatewayTask.Update(ctx, &entity.ModelGatewayTask{
-				SQLBaseDO:    beans.SQLBaseDO{Id: task.Id},
-				ExpendTokens: task.ExpendTokens,
-			})
-		}
-
-		// 3) 解析 + 校验
-		var parsed map[string]any
-		switch req.BuildType {
-		case public.BuildTypePrompt, public.BuildTypeNode:
-			parsed, err = util.ParseAndValidate(mapped, model)
-			if err == nil {
-				return parsed, nil
-			}
-		case public.BuildTypeStruct:
-			parsed = util.ParseStructResult(mapped, model.ResponseBody)
+		// 解析 + 校验（用构建模型的 RequiredFields）
+		parsed, err := util.ParseAndValidate(body, buildModel.RequiredFields)
+		if err == nil {
 			return parsed, nil
-		default:
-			return mapped, nil
 		}
+		lastErr = err
 
 		g.Log().Warningf(ctx, "[执行任务][解析失败] taskId=%s attempt=%d/%d err=%v", task.TaskID, attempt, maxRetry, err)
 
 		if attempt == maxRetry {
-			return nil, fmt.Errorf("JSON解析重试耗尽: %w", err)
+			return nil, fmt.Errorf("JSON解析重试耗尽: %w", lastErr)
 		}
 
-		// 4) 重新调模型（直接调，不走缓存）
+		// 重试：重新调模型
 		task.RetryCount++
 		_, _ = dao.ModelGatewayTask.Update(ctx, task)
-		rawData, callErr := InvokeModel(ctx, model, task.RequestPayload.Body)
 
+		reqBody := injectErrorMessage(task.RequestPayload.Body, lastErr)
+		rawData, callErr := InvokeModel(ctx, model, reqBody)
 		if callErr != nil {
 			g.Log().Warningf(ctx, "[执行任务][重调模型失败] taskId=%s attempt=%d/%d err=%v", task.TaskID, attempt, maxRetry, callErr)
 			continue
 		}
 
-		// 5) 解析原始响应，覆盖 body 进入下一轮
 		var rawResp map[string]any
-		if err = json.Unmarshal(rawData, &rawResp); err != nil {
+		if err := json.Unmarshal(rawData, &rawResp); err != nil {
 			g.Log().Warningf(ctx, "[执行任务][Unmarshal失败] taskId=%s err=%v", task.TaskID, err)
 			continue
 		}
-		body = rawResp
+		mapped, mapErr := util.MapResponsePayload(model.ResponseMapping, rawResp)
+		if mapErr != nil {
+			g.Log().Warningf(ctx, "[执行任务][映射失败] taskId=%s err=%v", task.TaskID, mapErr)
+			continue
+		}
+
+		// 计费
+		if len(model.BillingConfig) > 0 && len(task.BillingData) > 0 {
+			requestData := task.BillingData[0]
+			retryData := make(map[string]any)
+			for k, v := range requestData {
+				retryData[k] = v
+			}
+			responseData := util.ExtractResponseBilling(model.BillingConfig, mapped)
+			for k, v := range responseData {
+				retryData[k] = v
+			}
+			billingResult := util.CalculateBilling(model.BillingConfig, retryData)
+			if billingResult != nil {
+				task.BillingData = append(task.BillingData, billingResult)
+				totalFee := gconv.Float64(billingResult["total_fee"])
+				if totalFee > 0 {
+					for a := 0; a <= maxRetry; a++ {
+						errr := gateway.DeductBalance(util.AsyncCtx(ctx), task.TenantId, -totalFee)
+						if errr == nil {
+							break
+						}
+						g.Log().Warningf(ctx, "[handleOne] 扣除余额失败 taskId=%s attempt=%d/%d err=%v", task.TaskID, a, maxRetry, errr)
+					}
+				}
+			}
+
+			task.ExpendTokens += gconv.Int64(mapped[entity.TotalTokens])
+			_, _ = dao.ModelGatewayTask.Update(ctx, &entity.ModelGatewayTask{
+				SQLBaseDO:    beans.SQLBaseDO{Id: task.Id},
+				BillingData:  task.BillingData,
+				ExpendTokens: task.ExpendTokens,
+			})
+		}
+
+		body = mapped
 	}
 
 	return body, nil
 }
 
+// injectErrorMessage 将错误信息插入到最后一个 user 消息之前
+func injectErrorMessage(payload map[string]any, err error) map[string]any {
+	if err == nil {
+		return payload
+	}
+
+	messages, _ := payload["messages"].([]any)
+	if len(messages) == 0 {
+		return payload
+	}
+
+	errMsg := fmt.Sprintf("【上一轮输出错误，请修正】%s", err.Error())
+
+	// 找到最后一个 user 的位置
+	lastUserIdx := -1
+	for i := len(messages) - 1; i >= 0; i-- {
+		msg, ok := messages[i].(map[string]any)
+		if !ok {
+			continue
+		}
+		if gconv.String(msg["role"]) == "user" {
+			lastUserIdx = i
+			break
+		}
+	}
+
+	if lastUserIdx == -1 {
+		return payload
+	}
+
+	// 在最后一个 user 之前插入错误消息
+	errMsgObj := map[string]any{
+		"role":    "user",
+		"content": []map[string]any{{"type": "text", "text": errMsg}},
+	}
+
+	// 切片插入
+	messages = append(messages[:lastUserIdx], append([]any{errMsgObj}, messages[lastUserIdx:]...)...)
+	payload["messages"] = messages
+	return payload
+}
+
 // InvokeModel 调用模型服务，返回二进制结果
 // modelKey 用于覆盖/补充模型配置 head_msg（例如每次请求携带不同的 X-API-Key）
 func InvokeModel(ctx context.Context, model *entity.ModelGatewayModel, body map[string]any) ([]byte, error) {
-	// 1) 记录模型调用次数
-	_ = dao.ModelGatewayLogsStat.IncRequestCount(ctx, time.Now(), model.TenantId, model.Creator, model.ModelName)
-
-	// 2）请求参数映射：将标准 payload 按模型配置的 requestMapping 转为模型需要的格式
-	//—— 请求映射实际处理为提示词构建请求，因为有附加字段及其他字段的拼接。这里不方便做请求映射
-	//mappedPayload := util.ReverseMap(model.RequestMapping, payload)
+	//surplus, _ := gateway.GetTenantSurplus(ctx, model.TenantId)
+	//if surplus <= 0 {
+	//	return nil, fmt.Errorf("租户余额不足")
+	//}
 
 	// 3）构建请求 URL 和超时
 	baseURL := strings.TrimRight(model.BaseURL, "/")
@@ -392,13 +440,20 @@ func InvokeModel(ctx context.Context, model *entity.ModelGatewayModel, body map[
 				baseURL = baseURL + "?" + q.Encode()
 			}
 		}
-		req, err = http.NewRequestWithContext(ctx, http.MethodGet, baseURL, nil)
+		// 改用独立超时ctx，隔绝外层截止
+		reqCtx, reqCancel := context.WithTimeout(context.Background(), timeout)
+		defer reqCancel()
+		req, err = http.NewRequestWithContext(reqCtx, http.MethodGet, baseURL, nil)
+		//req, err = http.NewRequestWithContext(ctx, http.MethodGet, baseURL, nil)
 	default:
 		bodyBytes, err := json.Marshal(body)
 		if err != nil {
 			return nil, err
 		}
-		req, err = http.NewRequestWithContext(ctx, http.MethodPost, baseURL, bytes.NewReader(bodyBytes))
+		reqCtx, reqCancel := context.WithTimeout(context.Background(), timeout)
+		defer reqCancel()
+		req, err = http.NewRequestWithContext(reqCtx, http.MethodPost, baseURL, bytes.NewReader(bodyBytes))
+		//req, err = http.NewRequestWithContext(ctx, http.MethodPost, baseURL, bytes.NewReader(bodyBytes))
 	}
 
 	// 5）注入请求头：先模型静态配置，再动态 modelKey（后者可覆盖前者）
@@ -430,6 +485,11 @@ func InvokeModel(ctx context.Context, model *entity.ModelGatewayModel, body map[
 		msg := string(b)
 		return nil, fmt.Errorf("模型服务返回非2xx: %d, body=%s", resp.StatusCode, msg)
 	}
+
+	//
+
+	//
+	g.Log().Debugf(ctx, "[执行任务][模型调用成功] StatusCode=%v", resp.StatusCode)
 	return b, nil
 }
 
@@ -487,15 +547,13 @@ func InvokeModel(ctx context.Context, model *entity.ModelGatewayModel, body map[
 //		return mappedResponse, nil
 //	}
 
-// failTask 任务失败统一处理：更新数据库 + 释放排队 + 回调
+// failTask 任务失败统一处理
 func (w *asyncWorker) failTask(ctx context.Context, t *entity.ModelGatewayTask, startTime time.Time, errMsg string) {
 	t.State = 3
 	t.ErrorMsg = errMsg
 	t.DurationSeconds = int64(time.Since(startTime).Seconds())
-	_, err := dao.ModelGatewayTask.Update(ctx, t)
-	if err != nil {
-		g.Log().Warningf(ctx, "[执行任务][更新数据库失败] taskId=%s err=%v", t.TaskID, err)
-	}
-	queue.ReleaseQueueSlot(ctx, t.ModelName, t.TaskID)
-	go gateway.TriggerCallback(context.WithoutCancel(ctx), t)
+	concurrencyKey := fmt.Sprintf("model:concurrency:%s", t.ModelName)
+	g.Redis().Decr(ctx, concurrencyKey)
+	_, _ = dao.ModelGatewayTask.Update(ctx, t)        // 更新任务状态
+	go gateway.TriggerCallback(util.AsyncCtx(ctx), t) // 触发回调
 }

update.sql

@@ -230,4 +230,122 @@ COMMENT ON COLUMN model_gateway_logs_op.success IS '是否成功：1成功/0失
 COMMENT ON COLUMN model_gateway_logs_op.error_msg IS '错误信息（失败时）';
 COMMENT ON COLUMN model_gateway_logs_op.cost_ms IS '耗时（毫秒）';
 COMMENT ON COLUMN model_gateway_logs_op.request_payload IS '请求 JSON';
-COMMENT ON COLUMN model_gateway_logs_op.response_payload IS '响应 JSON';
+COMMENT ON COLUMN model_gateway_logs_op.response_payload IS '响应 JSON';
+
+-- =========================================================================================================================
+CREATE TABLE "public"."model_gateway_model_manage" (
+                                                       "id" int8 NOT NULL,
+                                                       "tenant_id" int8 NOT NULL DEFAULT 0,
+                                                       "creator" varchar(64) COLLATE "pg_catalog"."default" NOT NULL,
+                                                       "created_at" timestamp(6) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+                                                       "updater" varchar(64) COLLATE "pg_catalog"."default" NOT NULL,
+                                                       "updated_at" timestamp(6) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+                                                       "deleted_at" timestamp(6),
+                                                       "model_supplier" varchar(32) COLLATE "pg_catalog"."default" NOT NULL DEFAULT ''::character varying,
+                                                       "model_name" varchar(128) COLLATE "pg_catalog"."default" NOT NULL DEFAULT ''::character varying,
+                                                       "model_type" varchar(32) COLLATE "pg_catalog"."default" NOT NULL DEFAULT ''::character varying,
+                                                       "base_url" varchar(512) COLLATE "pg_catalog"."default" NOT NULL DEFAULT ''::character varying,
+                                                       "system_model" bool,
+                                                       "http_method" varchar(32) COLLATE "pg_catalog"."default" NOT NULL DEFAULT ''::character varying,
+                                                       "chat_model" bool,
+                                                       "response_type" int2 NOT NULL DEFAULT 0,
+                                                       "api_key" varchar(255) COLLATE "pg_catalog"."default" NOT NULL DEFAULT ''::character varying,
+                                                       "enabled" bool,
+                                                       "request_head_mapping" jsonb DEFAULT '{}'::jsonb,
+                                                       "request_body_mapping" jsonb DEFAULT '{}'::jsonb,
+                                                       "response_mapping" jsonb DEFAULT '{}'::jsonb,
+                                                       "max_concurrency" int4 NOT NULL DEFAULT 0,
+                                                       "token_mapping" jsonb,
+                                                       "async_task_mapping" jsonb,
+                                                       "token_predict_price" numeric(12,6) NOT NULL DEFAULT 0.000000,
+                                                       "max_tokens" int4 NOT NULL DEFAULT 0,
+                                                       "last_frame" varchar(512) COLLATE "pg_catalog"."default" NOT NULL DEFAULT ''::character varying,
+                                                       "response_body_mapping" jsonb DEFAULT '{}'::jsonb,
+                                                       "token_predict_price_unit" varchar(32) COLLATE "pg_catalog"."default",
+                                                       "max_duration" int4,
+                                                       CONSTRAINT "model_gateway_model_manage_pkey" PRIMARY KEY ("id")
+)
+;
+
+ALTER TABLE "public"."model_gateway_model_manage"
+    OWNER TO "postgres";
+
+CREATE INDEX "idx_model_manage_deleted_at" ON "public"."model_gateway_model_manage" USING btree (
+    "deleted_at" "pg_catalog"."timestamp_ops" ASC NULLS LAST
+    );
+
+CREATE INDEX "idx_model_manage_model_type" ON "public"."model_gateway_model_manage" USING btree (
+    "model_type" COLLATE "pg_catalog"."default" "pg_catalog"."text_ops" ASC NULLS LAST
+    );
+
+CREATE INDEX "idx_model_manage_response_type" ON "public"."model_gateway_model_manage" USING btree (
+    "response_type" "pg_catalog"."int2_ops" ASC NULLS LAST
+    );
+
+CREATE INDEX "idx_model_manage_supplier" ON "public"."model_gateway_model_manage" USING btree (
+    "model_supplier" COLLATE "pg_catalog"."default" "pg_catalog"."text_ops" ASC NULLS LAST
+    );
+
+CREATE INDEX "idx_model_manage_tenant_id" ON "public"."model_gateway_model_manage" USING btree (
+    "tenant_id" "pg_catalog"."int8_ops" ASC NULLS LAST
+    );
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."id" IS '主键ID';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."tenant_id" IS '租户ID';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."creator" IS '创建人';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."created_at" IS '创建时间';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."updater" IS '更新人';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."updated_at" IS '更新时间';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."deleted_at" IS '删除时间（软删）';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."model_supplier" IS '模型供应商';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."model_name" IS '模型名称';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."model_type" IS '模型类型';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."base_url" IS '模型地址';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."system_model" IS '是否系统模型';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."http_method" IS 'http请求方法';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."chat_model" IS '是否聊天模型';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."response_type" IS '返回类型：1同步，2异步，3流';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."api_key" IS '接口密钥';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."enabled" IS '是否启用';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."request_head_mapping" IS '请求头映射';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."request_body_mapping" IS '请求体映射';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."response_mapping" IS '响应映射';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."max_concurrency" IS '最大并发数';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."token_mapping" IS 'token映射';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."async_task_mapping" IS '异步任务映射';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."token_predict_price" IS '模型Token预估价格';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."max_tokens" IS '最大token数';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."last_frame" IS '视频尾帧图像地址';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."response_body_mapping" IS '响应主体映射';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."token_predict_price_unit" IS '模型token预估价格单位（秒，百万Token，千Token，字数）';
+
+COMMENT ON COLUMN "public"."model_gateway_model_manage"."max_duration" IS '最大时长（秒）';
+
+COMMENT ON TABLE "public"."model_gateway_model_manage" IS '模型管理表';