From b843bb8286d591b7ac2eb2bab98419e2171f9cbd Mon Sep 17 00:00:00 2001
From: CaIon <i@caion.me>
Date: Fri, 8 Aug 2025 10:43:07 +0800
Subject: [PATCH] feat: add support for gpt-5 models and adjust temperature
 settings

- Updated the model list to include various gpt-5 variants.
- Enhanced the ConvertOpenAIRequest function to handle gpt-5 model temperature settings based on specific model prefixes.
- Adjusted default cache and model ratios for new gpt-5 models.
---
 relay/channel/openai/adaptor.go      | 13 +++++++++++--
 relay/channel/openai/constant.go     |  3 +++
 setting/ratio_setting/cache_ratio.go | 10 ++++++++++
 setting/ratio_setting/model_ratio.go |  7 +++++++
 4 files changed, 31 insertions(+), 2 deletions(-)

diff --git a/relay/channel/openai/adaptor.go b/relay/channel/openai/adaptor.go
index 4d99a76c..01a5fa97 100644
--- a/relay/channel/openai/adaptor.go
+++ b/relay/channel/openai/adaptor.go
@@ -206,12 +206,21 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 			}
 		}
 	}
-	if strings.HasPrefix(request.Model, "o") {
+	if strings.HasPrefix(request.Model, "o") || strings.HasPrefix(request.Model, "gpt-5") {
 		if request.MaxCompletionTokens == 0 && request.MaxTokens != 0 {
 			request.MaxCompletionTokens = request.MaxTokens
 			request.MaxTokens = 0
 		}
-		request.Temperature = nil
+
+		if strings.HasPrefix(request.Model, "o") {
+			request.Temperature = nil
+		}
+
+		if strings.HasPrefix(request.Model, "gpt-5") {
+			if request.Model != "gpt-5-chat-latest" {
+				request.Temperature = nil
+			}
+		}
 
 		// 转换模型推理力度后缀
 		effort, originModel := parseReasoningEffortFromModelSuffix(request.Model)
diff --git a/relay/channel/openai/constant.go b/relay/channel/openai/constant.go
index c703e414..85a8ed17 100644
--- a/relay/channel/openai/constant.go
+++ b/relay/channel/openai/constant.go
@@ -18,6 +18,9 @@ var ModelList = []string{
 	"o3-mini-high", "o3-mini-2025-01-31-high",
 	"o3-mini-low", "o3-mini-2025-01-31-low",
 	"o3-mini-medium", "o3-mini-2025-01-31-medium",
+	"gpt-5", "gpt-5-2025-08-07", "gpt-5-chat-latest",
+	"gpt-5-mini", "gpt-5-mini-2025-08-07",
+	"gpt-5-nano", "gpt-5-nano-2025-08-07",
 	"o1", "o1-2024-12-17",
 	"gpt-4o-audio-preview", "gpt-4o-audio-preview-2024-10-01",
 	"gpt-4o-realtime-preview", "gpt-4o-realtime-preview-2024-10-01", "gpt-4o-realtime-preview-2024-12-17",
diff --git a/setting/ratio_setting/cache_ratio.go b/setting/ratio_setting/cache_ratio.go
index 8b87cb86..3f223bc3 100644
--- a/setting/ratio_setting/cache_ratio.go
+++ b/setting/ratio_setting/cache_ratio.go
@@ -25,6 +25,16 @@ var defaultCacheRatio = map[string]float64{
 	"gpt-4o-mini-realtime-preview":        0.5,
 	"gpt-4.5-preview":                     0.5,
 	"gpt-4.5-preview-2025-02-27":          0.5,
+	"gpt-4.1":                             0.25,
+	"gpt-4.1-mini":                        0.25,
+	"gpt-4.1-nano":                        0.25,
+	"gpt-5":                               0.1,
+	"gpt-5-2025-08-07":                    0.1,
+	"gpt-5-chat-latest":                   0.1,
+	"gpt-5-mini":                          0.1,
+	"gpt-5-mini-2025-08-07":               0.1,
+	"gpt-5-nano":                          0.1,
+	"gpt-5-nano-2025-08-07":               0.1,
 	"deepseek-chat":                       0.25,
 	"deepseek-reasoner":                   0.25,
 	"deepseek-coder":                      0.25,
diff --git a/setting/ratio_setting/model_ratio.go b/setting/ratio_setting/model_ratio.go
index c9a9521f..4a19895e 100644
--- a/setting/ratio_setting/model_ratio.go
+++ b/setting/ratio_setting/model_ratio.go
@@ -73,6 +73,13 @@ var defaultModelRatio = map[string]float64{
 	"gpt-4-turbo-2024-04-09":                  5, // $0.01 / 1K tokens
 	"gpt-4.5-preview":                         37.5,
 	"gpt-4.5-preview-2025-02-27":              37.5,
+	"gpt-5":                                   0.625,
+	"gpt-5-2025-08-07":                        0.625,
+	"gpt-5-chat-latest":                       0.625,
+	"gpt-5-mini":                              0.125,
+	"gpt-5-mini-2025-08-07":                   0.125,
+	"gpt-5-nano":                              0.025,
+	"gpt-5-nano-2025-08-07":                   0.025,
 	//"gpt-3.5-turbo-0301":           0.75, //deprecated
 	"gpt-3.5-turbo":          0.25,
 	"gpt-3.5-turbo-0613":     0.75,