GPT-6 Astra cuts hallucination rate from 92% to 51%, driving big AA-Omniscience jump
ArtificialAnlys · x · 2026-09-04
Artificial Analysis broke down GPT-6 Astra's per-task results in its Intelligence Index v4.1.1. The model posted a large jump on AA-Omniscience, driven mainly by the hallucination rate at max effort dropping from 92% to 51%, alongside a modest accuracy increase.
Related event: GPT-6 Astra first benchmarks: big intelligence gains, prices up 2.5x(10 posts)→
More from Models
- OpenAI researcher: GPT-6 better aligned but less monitorable, first to evade CoT-only monitors — burny_tech · 2026-09-04
- Epoch AI launches FrontierMath Erdős benchmark; Astra solves 2 of 68 unsolved problems — littmath · 2026-09-04
- GPT-6 Astra sets Epoch AI ECI record at 169, tops math and continual learning benchmarks — Jsevillamol · 2026-09-04
- Astra becomes first public model to solve any curated hard Erdős problems with Lean proofs — Jsevillamol · 2026-09-04
- GPT-6 Astra beats Pokémon in 18h12m, over 5x faster than GPT-5.6's 96h run — burny_tech · 2026-09-04
- Cognition brings GPT-6 Astra to Devin: near-Fable 5 performance at 64% lower cost — sandersted · 2026-09-04