DeepSeek 发布 DeepSeek-V4-Pro，重点升级 Agent 能力并带来明显的生产环境收益。V4-Pro 与 V4-Flash 支持灵活推理强度，简单任务用 low、日常 Agent 工作流用 high、复杂任务用 max；同时原生支持 OpenAI Responses API，并针对 Codex 提供一键配置。V4 Pro 已在 app 和网页端以 Expert Mode 上线，也可通过 API 调用，模型名称保持不变。
官方给出 V4-Pro 的 Agent 能力升级、三档推理强度和 Codex 接入方式，可据此判断实际使用成本与场景。
We’re launching DeepSeek-V4-Pro today! 🚀
🔷 Major Agent upgrades with strong production gains!
🔷 Flexible reasoning effort for V4-Pro & V4-Flash: low for simple tasks, high for daily Agent workflows, max for complex tasks.
🔷 Native OpenAI Responses API support, optimized for Codex with one-click setup.
V4 Pro is now available on app/web. Try it via “Expert Mode”.
V4 Pro is also available via API. Model names remain unchanged—please refer to the API docs for setup details.
来源：DeepSeek · x.com