Hope for add support for deepseek-v4

Before you asking

• I have tried the PDFMathTranslate-next and give feedback in PDFMathTranslate-next

Is your feature request related to a problem?

In Translate with different services function，the default model for DeepSeek is deepseek-chat which is a model without thinking. However, deepseek-v4 pro and flash has already published and in the future deepseek-chat will not be supported by DeepSeek . If we want to close think function，we must change the origin code. So we need add support for deepseek-v4 pro and flash.

Describe the solution you'd like

According to document of DeepSeek API. The new url is https://api.deepseek.com The way to close thinging for both pro and flash is:

response = client.chat.completions.create( model="deepseek-v4-pro",

...

extra_body={"thinking": {"type": "disabled"}} )

Additional context

deepseek-chat (将于 2026/07/24 弃用) deepseek-reasoner (将于 2026/07/24 弃用) (1) 默认思考开关为 enabled (2) 思考模式下，对普通请求，默认 effort 为 high；对一些复杂 Agent 类请求（如 Claude Code、OpenCode），effort 自动设置为 max (3) 思考模式下，出于兼容考虑 low、medium 会映射为 high, xhigh 会映射为 max 您在使用 OpenAI SDK 设置 thinking 参数时，需要将 thinking 参数传入 extra_body 中： response = client.chat.completions.create( model="deepseek-v4-pro",

...

reasoning_effort="high", extra_body={"thinking": {"type": "enabled"}} )