Speculation: DeepSeek V4.1 Pro may be a post-trained V4-Pro hitting a training wall
teortaxesTex · x · 2026-11-08
teortaxesTex speculates that if recent evals hold, DeepSeek's updated V4.1 Pro may just be a further post-trained V4-Pro — RL-driven reliability gains rather than real capability expansion — suggesting the same training wall that forced V4-line adjustments (and Kimi's K2.7–K2.8 fiddling). He adds it could be another routing incident, questioning eval reliability. Unverified personal speculation.
Related event: DeepSeek V4 Pro Suspected Silent Update Boosts Hard-Case Pass Rate(3 posts)→
More from Models
- Garbage in, garbage out: Opus makes bad videos from bad starting points — pramodk73 · 2026-11-08
- Counterintuitive: Opus 5.5 fails 2x more turns on max than on high, in every time bucket — tristanbob · 2026-11-08
- Thinking Machines serves Inkling 2 Small free on OpenCode for a limited time to gather feedback — clarejtbirch · 2026-11-08
- MaxxBench strips away benchmaxxing: Claude Opus 5.5 still tops at 74.2% — craterIII · 2026-11-08
- Claude wrote a wrong Pokémon landmark note and kept trusting it — Anthropic tested a memory repair pass — shoumikgoswami · 2026-11-08
- Sonnet 4.6 rates models on self-designed scales; author says 'self-report' caveats are performative — Sauers_ · 2026-11-08