DeepSeek V4 Pro (0813) Launches with Claude-Level Agent Performance
DeepSeek has officially released its V4 Pro large language model, designated DeepSeek‑V4‑Pro‑0813, with unchanged API pricing, a noticeable shift in chain‑of‑thought behavior, and benchmark results that put its agent capabilities on par with top models like Claude Fable 5, while recent V4 Flash users report performance drops and potential price hikes.
On Wednesday evening DeepSeek unexpectedly released the official version of its V4 Pro large language model, identified as DeepSeek‑V4‑Pro‑0813.
The earliest update appears in the DeepSeek API documentation, confirming the new model name and noting that the API pricing has not been increased.
Members of the AI community who already use large‑model APIs report that the V4 Pro’s chain‑of‑thought style has changed compared with previous versions.
A performance chart circulating online shows that V4 Pro’s agent‑related benchmark scores are nearly on par with leading models such as Claude Fable 5.
Earlier, the release of DeepSeek V4 Flash led to widespread usage and caused the API to suffer performance degradation due to insufficient inference compute; the company is reportedly considering a price increase.
Questions remain whether the V4 Pro official release can meet users’ expectations for cost‑performance, and whether further surprises, such as the upcoming Harness release, will arrive soon.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.