DeepSeek 模型升级至 V4 系列，使用官方 API 命名体系

- 新增 deepseek-v4-pro（专家模式）和 deepseek-v4-flash（快速模式）
- 模型通过后缀 -think / -search 独立控制思考和搜索
- chat completion 新增 model_type / parent_message_id / preempt 字段
- 保留 deepseek-chat / deepseek-reasoner 兼容旧名
- 移除已下线的 DeepSeek-V3.2 / R1 / Search / R1-Search

Co-Authored-By: DeepSeek V4 Pro <noreply@deepseek.com>

Author: bennylii

README.md

@@ -49,7 +49,7 @@
 
 | Provider         | Auth Type     | OAuth | Models                                                                          |
 | ---------------- | ------------- | ----- | ------------------------------------------------------------------------------- |
-| DeepSeek         | User Token    | Yes   | DeepSeek-V3.2                                                                   |
+| DeepSeek         | User Token    | Yes   | deepseek-v4-pro, deepseek-v4-flash                                                                   |
 | GLM              | Refresh Token | Yes   | GLM-5                                                                           |
 | Kimi             | JWT Token     | Yes   | kimi-k2.5                                                                       |
 | MiniMax          | JWT Token     | Yes   | MiniMax-M2.5                                                                    |
@@ -145,7 +145,7 @@ client = OpenAI(
 )
 
 response = client.chat.completions.create(
-    model="DeepSeek-V3.2",
+    model="deepseek-v4-pro",
     messages=[
         {"role": "user", "content": "Hello, who are you?"}
     ]

README_CN.md

@@ -49,7 +49,7 @@
 
 | 服务商           | 认证类型          | OAuth | 模型                                                                              |
 | ------------- | ------------- | ----- | ------------------------------------------------------------------------------- |
-| DeepSeek      | User Token    | 是     | DeepSeek-V3.2                                                                   |
+| DeepSeek      | User Token    | 是     | deepseek-v4-pro, deepseek-v4-flash                                                                   |
 | GLM           | Refresh Token | 是     | GLM-5                                                                           |
 | Kimi          | JWT Token     | 是     | kimi-k2.5                                                                       |
 | MiniMax       | JWT Token     | 是     | MiniMax-M2.5                                                                    |
@@ -145,7 +145,7 @@ client = OpenAI(
 )
 
 response = client.chat.completions.create(
-    model="DeepSeek-V3.2",
+    model="deepseek-v4-pro",
     messages=[
         {"role": "user", "content": "你好，你是谁？"}
     ]