Merge pull request #1523 from RedwindA/refactor/improve-modelName-wildcard

fix: 修复通配符处理对判断可用渠道的破坏
This commit is contained in:
Calcium-Ion
2025-08-08 10:22:47 +08:00
committed by GitHub
2 changed files with 18 additions and 7 deletions

View File

@@ -129,8 +129,6 @@ func CacheGetRandomSatisfiedChannel(c *gin.Context, group string, model string,
}
func getRandomSatisfiedChannel(group string, model string, retry int) (*Channel, error) {
model = ratio_setting.FormatMatchingModelName(model)
// if memory cache is disabled, get channel directly from database
if !common.MemoryCacheEnabled {
return GetRandomSatisfiedChannel(group, model, retry)
@@ -138,8 +136,16 @@ func getRandomSatisfiedChannel(group string, model string, retry int) (*Channel,
channelSyncLock.RLock()
defer channelSyncLock.RUnlock()
// First, try to find channels with the exact model name.
channels := group2model2channels[group][model]
// If no channels found, try to find channels with the normalized model name.
if len(channels) == 0 {
normalizedModel := ratio_setting.FormatMatchingModelName(model)
channels = group2model2channels[group][normalizedModel]
}
if len(channels) == 0 {
return nil, nil
}

View File

@@ -150,6 +150,7 @@ var defaultModelRatio = map[string]float64{
"gemini-2.5-flash-preview-05-20-nothinking": 0.075,
"gemini-2.5-flash-thinking-*": 0.075, // 用于为后续所有2.5 flash thinking budget 模型设置默认倍率
"gemini-2.5-pro-thinking-*": 0.625, // 用于为后续所有2.5 pro thinking budget 模型设置默认倍率
"gemini-2.5-flash-lite-preview-thinking-*": 0.05,
"gemini-2.5-flash-lite-preview-06-17": 0.05,
"gemini-2.5-flash": 0.15,
"text-embedding-004": 0.001,
@@ -503,9 +504,6 @@ func getHardcodedCompletionModelRatio(name string) (float64, bool) {
return 3.5 / 0.15, false
}
if strings.HasPrefix(name, "gemini-2.5-flash-lite") {
if strings.HasPrefix(name, "gemini-2.5-flash-lite-preview") {
return 4, false
}
return 4, false
}
return 2.5 / 0.3, true
@@ -657,8 +655,15 @@ func GetCompletionRatioCopy() map[string]float64 {
// 转换模型名,减少渠道必须配置各种带参数模型
func FormatMatchingModelName(name string) string {
name = handleThinkingBudgetModel(name, "gemini-2.5-flash", "gemini-2.5-flash-thinking-*")
name = handleThinkingBudgetModel(name, "gemini-2.5-pro", "gemini-2.5-pro-thinking-*")
if strings.HasPrefix(name, "gemini-2.5-flash-lite") {
name = handleThinkingBudgetModel(name, "gemini-2.5-flash-lite", "gemini-2.5-flash-lite-thinking-*")
} else if strings.HasPrefix(name, "gemini-2.5-flash") {
name = handleThinkingBudgetModel(name, "gemini-2.5-flash", "gemini-2.5-flash-thinking-*")
} else if strings.HasPrefix(name, "gemini-2.5-pro") {
name = handleThinkingBudgetModel(name, "gemini-2.5-pro", "gemini-2.5-pro-thinking-*")
}
if strings.HasPrefix(name, "gpt-4-gizmo") {
name = "gpt-4-gizmo-*"
}