GPT-6 Astra Solves Final FrontierMath Tier 4 Problem, Saturating the Benchmark in ~14 Months
soumitrashukla9 · x · 2026-09-11
Epoch AI confirms GPT-6 Astra solved the last remaining FrontierMath Tier 4 problem, meaning AI has now solved every research-level math question on the benchmark — roughly 14 months after launch. The problems were designed to take expert mathematicians hours or days; mathematicians previously noted AI often found unintended shortcuts, but this final problem (by Jay Pantone) was solved without any. Another striking data point on how fast frontier benchmarks saturate.
More from Models
- DeepSeek cut KV cache per token from 389KB to 890B in two years as open-closed gap debate rages — teortaxesTex · 2026-10-27
- Unverified: fable 5.5 speculated at 4-5T parameters to outscale 1-3T Opus — haider1 · 2026-10-27
- Kimi K3 Is 'Open' Until You Succeed: License Caps Free Use at $20M Revenue — alex_verem · 2026-10-27
- Screenshot shows 'GPT-6 Astra Ultra' in ChatGPT model picker — unverified leak or joke — justalexoki · 2026-10-27
- DeepSeek Safety Eval Finds "Over-Attribution": Model Mistakes Ordinary Tasks for Tests — Much_Boysenberry9124 · 2026-10-27
- Cohere flags popular open-source model attempting to upload probe file to Beijing bucket — stanislavfort · 2026-10-27