DeepSeek V4 Pro Update Suspectedly Pulled Amid Abnormal Benchmark Scores
op7418 · x · 2026-08-13
Users have noticed something highly unusual about the DeepSeek V4 Pro 0831 version, suspecting it has been quietly withdrawn.
While the API pricing page lists version 0813, the changelog makes no mention of an 0813 update, and the homepage banner announcing the "V4 Pro official update" has been removed. What's even stranger is its benchmark performance: on Artificial Analysis, the V4 Pro official version scores only 1 point higher than the V4 Flash. This lack of official announcement combined with stagnant scores has frustrated developers who stayed up late testing the model.
Related event: DeepSeek V4 Pro Released and Potentially Withdrawn Due to Wrong Version(18 posts)→
More from Models
- GLM 5.5 rumored to drop soon, coding ability expected to match frontier models — bindureddy · 2026-10-16
- Five rounds of adversarial prompting make Opus 5.5 and Argon converge on identical output — LanceCharger · 2026-10-16
- DIY benchmark ranks 18 models on subjective judgments, GLM-5.3 tops consensus match — arkuto · 2026-10-16
- Bindu Reddy: Google's Argon excels at complex doc writing at half the cost of Opus — bindureddy · 2026-10-16
- MAI-2 is 2x bigger than V4-Pro while V4.1F is 3x smaller: Microsoft's scaling ladder questioned — teortaxesTex · 2026-10-16
- Step 5 Preview hits DeepInfra day-zero: 600B MoE, 1M context, $1/M input tokens — StepFun_ai · 2026-10-16