MathArena adds October BrokenArXiv: GPT-6.1 Sol leads but drifts down monthly, Opus 5.5 rock solid
teortaxesTex · x · 2026-11-09
MathArena added October samples of BrokenArXiv and ArXivMath. GPT-6.1 Sol leads on both benchmarks, while Gemini 4 Argon does well on ArXivMath but terribly on BrokenArXiv.
teortaxesTex notes that even the strongest models regress on newer samples month by month: GPT-6.1 Sol fell from 61 to 56 to 50 on BrokenArXiv (1017) and 86→83→80 on September samples — more robust, but still declining. The exception is Claude Opus 5.5, which stays rock solid, suggesting Anthropic didn't benchmaxx against these sets.
More from Models
- Polymarket odds of "Fable" launching today surge to 70% — spreadlove5683 · 2026-11-10
- Congress returns to session with expected focus on OpenAI rogue model incidents — peterwildeford · 2026-11-10
- "Benchmarks rule out the bad models — the good ones I test on my own work" — ChemicalExcellent463 · 2026-11-10
- Elon Musk amplifies Grok Bot feedback thread as team works on token efficiency fixes — elonmusk · 2026-11-10
- PostTrainBench v1.2: Gemini 4 Argon Scores 40.4% (#5), Muse Wonder 24.2% (#17) — maksym_andr · 2026-11-09
- Users report Opus 5.5 back to normal, suspect Anthropic silently reverted the degradation — ryunuck · 2026-11-09