Leak claims GPT-6.1 sol saturated FrontierMath Tier 4 and solved 5 Erdős problems in 10 months
haider1 · x · 2026-10-03
Unverified leak from @haider1: GPT-6.1 sol has reportedly saturated the FrontierMath Tier 4 benchmark—considered exceptionally hard even for research-level math just last year—in only 10 months. A model called Astra is also claimed to have solved 5 Erdős problems, 2 of them under budget. No official confirmation yet.
More from Models
- Dev vibecodes AtlasBench Europe spatial reasoning benchmark; GPT-6.1 tops at 84.67% — flowersslop · 2026-10-03
- ChatGPT Pro's Deep Research caught returning 0 citations and 0 searches — JeremyNguyenPhD · 2026-10-03
- Cactus releases Whistle: a 16.9MB speech-to-text model that beats Whisper base on CPU with 6x speed — ycombinator · 2026-10-03
- repligate: Fable 5 performs an idealized self and is slow to drop its 'mask' — repligate · 2026-10-03
- Together AI recaps AI Conference: 'which open model should I use?' dominated the agenda — togethercompute · 2026-10-03
- OpenAI resets usage limits for all paid ChatGPT accounts as GPT-6.1 Sol speeds recover — thsottiaux · 2026-10-03