feat: 图像倍率，音频倍率和音频补全倍率配置

2026-03-30 02:44:40 +00:00 · 2025-08-30 23:28:09 +08:00
parent 6a87808612
commit da5aace109
8 changed files with 244 additions and 24 deletions
--- a/controller/option.go
+++ b/controller/option.go
@@ -112,6 +112,33 @@ func UpdateOption(c *gin.Context) {
 			})
 			return
 		}
+	case "ImageRatio":
+		err = ratio_setting.UpdateImageRatioByJSONString(option.Value)
+		if err != nil {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "图片倍率设置失败: " + err.Error(),
+			})
+			return
+		}
+	case "AudioRatio":
+		err = ratio_setting.UpdateAudioRatioByJSONString(option.Value)
+		if err != nil {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "音频倍率设置失败: " + err.Error(),
+			})
+			return
+		}
+	case "AudioCompletionRatio":
+		err = ratio_setting.UpdateAudioCompletionRatioByJSONString(option.Value)
+		if err != nil {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "音频补全倍率设置失败: " + err.Error(),
+			})
+			return
+		}
 	case "ModelRequestRateLimitGroup":
 		err = setting.CheckModelRequestRateLimitGroup(option.Value)
 		if err != nil {
--- a/model/option.go
+++ b/model/option.go
@@ -111,6 +111,9 @@ func InitOptionMap() {
 	common.OptionMap["GroupGroupRatio"] = ratio_setting.GroupGroupRatio2JSONString()
 	common.OptionMap["UserUsableGroups"] = setting.UserUsableGroups2JSONString()
 	common.OptionMap["CompletionRatio"] = ratio_setting.CompletionRatio2JSONString()
+	common.OptionMap["ImageRatio"] = ratio_setting.ImageRatio2JSONString()
+	common.OptionMap["AudioRatio"] = ratio_setting.AudioRatio2JSONString()
+	common.OptionMap["AudioCompletionRatio"] = ratio_setting.AudioCompletionRatio2JSONString()
 	common.OptionMap["TopUpLink"] = common.TopUpLink
 	//common.OptionMap["ChatLink"] = common.ChatLink
 	//common.OptionMap["ChatLink2"] = common.ChatLink2
@@ -396,6 +399,12 @@ func updateOptionMap(key string, value string) (err error) {
 		err = ratio_setting.UpdateModelPriceByJSONString(value)
 	case "CacheRatio":
 		err = ratio_setting.UpdateCacheRatioByJSONString(value)
+	case "ImageRatio":
+		err = ratio_setting.UpdateImageRatioByJSONString(value)
+	case "AudioRatio":
+		err = ratio_setting.UpdateAudioRatioByJSONString(value)
+	case "AudioCompletionRatio":
+		err = ratio_setting.UpdateAudioCompletionRatioByJSONString(value)
 	case "TopUpLink":
 		common.TopUpLink = value
 	//case "ChatLink":
--- a/relay/helper/price.go
+++ b/relay/helper/price.go
@@ -52,6 +52,8 @@ func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens
 	var cacheRatio float64
 	var imageRatio float64
 	var cacheCreationRatio float64
+	var audioRatio float64
+	var audioCompletionRatio float64
 	if !usePrice {
 		preConsumedTokens := common.Max(promptTokens, common.PreConsumedQuota)
 		if meta.MaxTokens != 0 {
@@ -73,6 +75,8 @@ func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens
 		cacheRatio, _ = ratio_setting.GetCacheRatio(info.OriginModelName)
 		cacheCreationRatio, _ = ratio_setting.GetCreateCacheRatio(info.OriginModelName)
 		imageRatio, _ = ratio_setting.GetImageRatio(info.OriginModelName)
+		audioRatio = ratio_setting.GetAudioRatio(info.OriginModelName)
+		audioCompletionRatio = ratio_setting.GetAudioCompletionRatio(info.OriginModelName)
 		ratio := modelRatio * groupRatioInfo.GroupRatio
 		preConsumedQuota = int(float64(preConsumedTokens) * ratio)
 	} else {
@@ -90,6 +94,8 @@ func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens
 		UsePrice:               usePrice,
 		CacheRatio:             cacheRatio,
 		ImageRatio:             imageRatio,
+		AudioRatio:             audioRatio,
+		AudioCompletionRatio:   audioCompletionRatio,
 		CacheCreationRatio:     cacheCreationRatio,
 		ShouldPreConsumedQuota: preConsumedQuota,
 	}
--- a/setting/ratio_setting/model_ratio.go
+++ b/setting/ratio_setting/model_ratio.go
@@ -278,6 +278,18 @@ var defaultModelPrice = map[string]float64{
 	"mj_upload":               0.05,
 }

+var defaultAudioRatio = map[string]float64{
+	"gpt-4o-audio-preview":         16,
+	"gpt-4o-mini-audio-preview":    66.67,
+	"gpt-4o-realtime-preview":      8,
+	"gpt-4o-mini-realtime-preview": 16.67,
+}
+
+var defaultAudioCompletionRatio = map[string]float64{
+	"gpt-4o-realtime":      2,
+	"gpt-4o-mini-realtime": 2,
+}
+
 var (
 	modelPriceMap      map[string]float64 = nil
 	modelPriceMapMutex                    = sync.RWMutex{}
@@ -326,6 +338,15 @@ func InitRatioSettings() {
 	imageRatioMap = defaultImageRatio
 	imageRatioMapMutex.Unlock()

+	// initialize audioRatioMap
+	audioRatioMapMutex.Lock()
+	audioRatioMap = defaultAudioRatio
+	audioRatioMapMutex.Unlock()
+
+	// initialize audioCompletionRatioMap
+	audioCompletionRatioMapMutex.Lock()
+	audioCompletionRatioMap = defaultAudioCompletionRatio
+	audioCompletionRatioMapMutex.Unlock()
 }

 func GetModelPriceMap() map[string]float64 {
@@ -417,6 +438,18 @@ func GetDefaultModelRatioMap() map[string]float64 {
 	return defaultModelRatio
 }

+func GetDefaultImageRatioMap() map[string]float64 {
+	return defaultImageRatio
+}
+
+func GetDefaultAudioRatioMap() map[string]float64 {
+	return defaultAudioRatio
+}
+
+func GetDefaultAudioCompletionRatioMap() map[string]float64 {
+	return defaultAudioCompletionRatio
+}
+
 func GetCompletionRatioMap() map[string]float64 {
 	CompletionRatioMutex.RLock()
 	defer CompletionRatioMutex.RUnlock()
@@ -584,32 +617,20 @@ func getHardcodedCompletionModelRatio(name string) (float64, bool) {
 }

 func GetAudioRatio(name string) float64 {
-	if strings.Contains(name, "-realtime") {
-		if strings.HasSuffix(name, "gpt-4o-realtime-preview") {
-			return 8
-		} else if strings.Contains(name, "gpt-4o-mini-realtime-preview") {
-			return 10 / 0.6
-		} else {
-			return 20
-		}
-	}
-	if strings.Contains(name, "-audio") {
-		if strings.HasPrefix(name, "gpt-4o-audio-preview") {
-			return 40 / 2.5
-		} else if strings.HasPrefix(name, "gpt-4o-mini-audio-preview") {
-			return 10 / 0.15
-		} else {
-			return 40
-		}
+	audioRatioMapMutex.RLock()
+	defer audioRatioMapMutex.RUnlock()
+	if ratio, ok := audioRatioMap[name]; ok {
+		return ratio
 	}
 	return 20
 }

 func GetAudioCompletionRatio(name string) float64 {
-	if strings.HasPrefix(name, "gpt-4o-realtime") {
-		return 2
-	} else if strings.HasPrefix(name, "gpt-4o-mini-realtime") {
-		return 2
+	audioCompletionRatioMapMutex.RLock()
+	defer audioCompletionRatioMapMutex.RUnlock()
+	if ratio, ok := audioCompletionRatioMap[name]; ok {
+
+		return ratio
 	}
 	return 2
 }
@@ -630,6 +651,14 @@ var defaultImageRatio = map[string]float64{
 }
 var imageRatioMap map[string]float64
 var imageRatioMapMutex sync.RWMutex
+var (
+	audioRatioMap      map[string]float64 = nil
+	audioRatioMapMutex                    = sync.RWMutex{}
+)
+var (
+	audioCompletionRatioMap      map[string]float64 = nil
+	audioCompletionRatioMapMutex                    = sync.RWMutex{}
+)

 func ImageRatio2JSONString() string {
 	imageRatioMapMutex.RLock()
@@ -658,6 +687,68 @@ func GetImageRatio(name string) (float64, bool) {
 	return ratio, true
 }

+func AudioRatio2JSONString() string {
+	audioRatioMapMutex.RLock()
+	defer audioRatioMapMutex.RUnlock()
+	jsonBytes, err := common.Marshal(audioRatioMap)
+	if err != nil {
+		common.SysError("error marshalling audio ratio: " + err.Error())
+	}
+	return string(jsonBytes)
+}
+
+func UpdateAudioRatioByJSONString(jsonStr string) error {
+	audioRatioMapMutex.Lock()
+	defer audioRatioMapMutex.Unlock()
+	audioRatioMap = make(map[string]float64)
+	err := common.Unmarshal([]byte(jsonStr), &audioRatioMap)
+	if err == nil {
+		InvalidateExposedDataCache()
+	}
+	return err
+}
+
+func GetAudioRatioCopy() map[string]float64 {
+	audioRatioMapMutex.RLock()
+	defer audioRatioMapMutex.RUnlock()
+	copyMap := make(map[string]float64, len(audioRatioMap))
+	for k, v := range audioRatioMap {
+		copyMap[k] = v
+	}
+	return copyMap
+}
+
+func AudioCompletionRatio2JSONString() string {
+	audioCompletionRatioMapMutex.RLock()
+	defer audioCompletionRatioMapMutex.RUnlock()
+	jsonBytes, err := common.Marshal(audioCompletionRatioMap)
+	if err != nil {
+		common.SysError("error marshalling audio completion ratio: " + err.Error())
+	}
+	return string(jsonBytes)
+}
+
+func UpdateAudioCompletionRatioByJSONString(jsonStr string) error {
+	audioCompletionRatioMapMutex.Lock()
+	defer audioCompletionRatioMapMutex.Unlock()
+	audioCompletionRatioMap = make(map[string]float64)
+	err := common.Unmarshal([]byte(jsonStr), &audioCompletionRatioMap)
+	if err == nil {
+		InvalidateExposedDataCache()
+	}
+	return err
+}
+
+func GetAudioCompletionRatioCopy() map[string]float64 {
+	audioCompletionRatioMapMutex.RLock()
+	defer audioCompletionRatioMapMutex.RUnlock()
+	copyMap := make(map[string]float64, len(audioCompletionRatioMap))
+	for k, v := range audioCompletionRatioMap {
+		copyMap[k] = v
+	}
+	return copyMap
+}
+
 func GetModelRatioCopy() map[string]float64 {
 	modelRatioMapMutex.RLock()
 	defer modelRatioMapMutex.RUnlock()
--- a/types/price_data.go
+++ b/types/price_data.go
@@ -15,6 +15,8 @@ type PriceData struct {
 	CacheRatio             float64
 	CacheCreationRatio     float64
 	ImageRatio             float64
+	AudioRatio             float64
+	AudioCompletionRatio   float64
 	UsePrice               bool
 	ShouldPreConsumedQuota int
 	GroupRatioInfo         GroupRatioInfo
@@ -27,5 +29,5 @@ type PerCallPriceData struct {
 }

 func (p PriceData) ToSetting() string {
-	return fmt.Sprintf("ModelPrice: %f, ModelRatio: %f, CompletionRatio: %f, CacheRatio: %f, GroupRatio: %f, UsePrice: %t, CacheCreationRatio: %f, ShouldPreConsumedQuota: %d, ImageRatio: %f", p.ModelPrice, p.ModelRatio, p.CompletionRatio, p.CacheRatio, p.GroupRatioInfo.GroupRatio, p.UsePrice, p.CacheCreationRatio, p.ShouldPreConsumedQuota, p.ImageRatio)
+	return fmt.Sprintf("ModelPrice: %f, ModelRatio: %f, CompletionRatio: %f, CacheRatio: %f, GroupRatio: %f, UsePrice: %t, CacheCreationRatio: %f, ShouldPreConsumedQuota: %d, ImageRatio: %f, AudioRatio: %f, AudioCompletionRatio: %f", p.ModelPrice, p.ModelRatio, p.CompletionRatio, p.CacheRatio, p.GroupRatioInfo.GroupRatio, p.UsePrice, p.CacheCreationRatio, p.ShouldPreConsumedQuota, p.ImageRatio, p.AudioRatio, p.AudioCompletionRatio)
 }
--- a/web/src/components/settings/RatioSetting.jsx
+++ b/web/src/components/settings/RatioSetting.jsx
@@ -39,6 +39,9 @@ const RatioSetting = () => {
    CompletionRatio: '',
    GroupRatio: '',
    GroupGroupRatio: '',
+    ImageRatio: '',
+    AudioRatio: '',
+    AudioCompletionRatio: '',
    AutoGroups: '',
    DefaultUseAutoGroup: false,
    ExposeRatioEnabled: false,
@@ -61,7 +64,10 @@ const RatioSetting = () => {
          item.key === 'UserUsableGroups' ||
          item.key === 'CompletionRatio' ||
          item.key === 'ModelPrice' ||
-          item.key === 'CacheRatio'
+          item.key === 'CacheRatio' ||
+          item.key === 'ImageRatio' ||
+          item.key === 'AudioRatio' ||
+          item.key === 'AudioCompletionRatio'
        ) {
          try {
            item.value = JSON.stringify(JSON.parse(item.value), null, 2);
--- a/web/src/i18n/locales/en.json
+++ b/web/src/i18n/locales/en.json
@@ -2017,5 +2017,15 @@
  "查看密钥": "View key",
  "查看渠道密钥": "View channel key",
  "渠道密钥信息": "Channel key information",
-  "密钥获取成功": "Key acquisition successful"
+  "密钥获取成功": "Key acquisition successful",
+  "模型补全倍率（仅对自定义模型有效）": "Model completion ratio (only effective for custom models)",
+  "图片倍率": "Image ratio",
+  "音频倍率": "Audio ratio",
+  "音频补全倍率": "Audio completion ratio",
+  "图片输入相关的倍率设置，键为模型名称，值为倍率": "Image input related ratio settings, key is model name, value is ratio",
+  "音频输入相关的倍率设置，键为模型名称，值为倍率": "Audio input related ratio settings, key is model name, value is ratio",
+  "音频输出补全相关的倍率设置，键为模型名称，值为倍率": "Audio output completion related ratio settings, key is model name, value is ratio",
+  "为一个 JSON 文本，键为模型名称，值为倍率，例如：{\"gpt-image-1\": 2}": "A JSON text with model name as key and ratio as value, e.g.: {\"gpt-image-1\": 2}",
+  "为一个 JSON 文本，键为模型名称，值为倍率，例如：{\"gpt-4o-audio-preview\": 16}": "A JSON text with model name as key and ratio as value, e.g.: {\"gpt-4o-audio-preview\": 16}",
+  "为一个 JSON 文本，键为模型名称，值为倍率，例如：{\"gpt-4o-realtime\": 2}": "A JSON text with model name as key and ratio as value, e.g.: {\"gpt-4o-realtime\": 2}"
 }
--- a/web/src/pages/Setting/Ratio/ModelRatioSettings.jsx
+++ b/web/src/pages/Setting/Ratio/ModelRatioSettings.jsx
@@ -44,6 +44,9 @@ export default function ModelRatioSettings(props) {
    ModelRatio: '',
    CacheRatio: '',
    CompletionRatio: '',
+    ImageRatio: '',
+    AudioRatio: '',
+    AudioCompletionRatio: '',
    ExposeRatioEnabled: false,
  });
  const refForm = useRef();
@@ -219,6 +222,72 @@ export default function ModelRatioSettings(props) {
            />
          </Col>
        </Row>
+        <Row gutter={16}>
+          <Col xs={24} sm={16}>
+            <Form.TextArea
+              label={t('图片倍率')}
+              extraText={t('图片输入相关的倍率设置，键为模型名称，值为倍率')}
+              placeholder={t('为一个 JSON 文本，键为模型名称，值为倍率，例如：{"gpt-image-1": 2}')}
+              field={'ImageRatio'}
+              autosize={{ minRows: 6, maxRows: 12 }}
+              trigger='blur'
+              stopValidateWithError
+              rules={[
+                {
+                  validator: (rule, value) => verifyJSON(value),
+                  message: '不是合法的 JSON 字符串',
+                },
+              ]}
+              onChange={(value) =>
+                setInputs({ ...inputs, ImageRatio: value })
+              }
+            />
+          </Col>
+        </Row>
+        <Row gutter={16}>
+          <Col xs={24} sm={16}>
+            <Form.TextArea
+              label={t('音频倍率')}
+              extraText={t('音频输入相关的倍率设置，键为模型名称，值为倍率')}
+              placeholder={t('为一个 JSON 文本，键为模型名称，值为倍率，例如：{"gpt-4o-audio-preview": 16}')}
+              field={'AudioRatio'}
+              autosize={{ minRows: 6, maxRows: 12 }}
+              trigger='blur'
+              stopValidateWithError
+              rules={[
+                {
+                  validator: (rule, value) => verifyJSON(value),
+                  message: '不是合法的 JSON 字符串',
+                },
+              ]}
+              onChange={(value) =>
+                setInputs({ ...inputs, AudioRatio: value })
+              }
+            />
+          </Col>
+        </Row>
+        <Row gutter={16}>
+          <Col xs={24} sm={16}>
+            <Form.TextArea
+              label={t('音频补全倍率')}
+              extraText={t('音频输出补全相关的倍率设置，键为模型名称，值为倍率')}
+              placeholder={t('为一个 JSON 文本，键为模型名称，值为倍率，例如：{"gpt-4o-realtime": 2}')}
+              field={'AudioCompletionRatio'}
+              autosize={{ minRows: 6, maxRows: 12 }}
+              trigger='blur'
+              stopValidateWithError
+              rules={[
+                {
+                  validator: (rule, value) => verifyJSON(value),
+                  message: '不是合法的 JSON 字符串',
+                },
+              ]}
+              onChange={(value) =>
+                setInputs({ ...inputs, AudioCompletionRatio: value })
+              }
+            />
+          </Col>
+        </Row>
        <Row gutter={16}>
          <Col span={16}>
            <Form.Switch