Chinese artificial intelligence startup DeepSeek released on late Wednesday the official API version of its V4 Pro model, delivering a major upgrade in AI agent capabilities, with benchmark results showing the model closing in on — and in some tests surpassing — Claude Fable 5.

DeepSeek launched social media accounts and posted job listings for a new Harness team building AI coding agents to rival Claude Code, Copilot,

The April preview got benchmarked to death and landed 18 points behind Anthropic’s flagship. DeepSeek’s own numbers tell a different story.