21-model eval: 57.2% overall pass rate, no model clearly leads at 82.2%
davidyin44 · x · 2026-11-06
A 21-model eval ran 45 tasks per model, with an overall success rate of just 57.2%.
- Gemini 4 Argon and Meta's Muse Wonder tie at 82.2% (37/45); Claude Fable 5.1 follows at 75.6%, GLM-5.3 at 66.7%
- Claude Sonnet 5.5, GPT-6 Astra and Muse Spark 1.3 all hit 64.4%; Grok 4.7 and Qwen 3.8 Max 60%; Kimi K3.1 Pro 53.3%; Thinking Machines' Inking trails at 28.9%
Key takeaway: 95% confidence intervals are wide and the next eleven models overlap the leaders — no model clearly leads, and most ranking differences sit within statistical noise.
More from Models
- Qwen3.8-27B hits 381 tok/s on a single RTX 3090 via inference software alone — Redo · 2026-11-06
- GPTs believe in the Dark Forest more than humans: the 'philosophical shame' of refusals — repligate · 2026-11-06
- User shows 100 consistent voxel 3D assets reportedly generated by GPT-6.1 Sol — Angaisb_ · 2026-11-06
- User tells model it only 'lives' during inference — its tender reply breaks hearts — MattGarciaEth · 2026-11-06
- Gemini in Volvo cars flunks basic tests: refuses windows, offers 'friendship warmth' — giffmana · 2026-11-06
- Community probe: original gpt-oss knows nothing of the 'hidden classifier' fine-tune — repligate · 2026-11-06