GPT-6 Astra Beats Sol max 2x on ARC v3 at Low Reasoning

Tests shared by Mike Knoop show GPT-6 Astra scoring about twice Sol max's accuracy on ARC v3 at low reasoning settings, using reasoning tokens on only 27% of actions, suggesting a major leap in implicit reasoning.

2026-09-05 ~ 2026-09-05 · 2 related posts

• GPT-6 Astra uses reasoning tokens in only 27% of ARC-AGI-3 actions, hinting at a leap in opaque reasoning — mhmazur · 2026-09-05
• ARC v3: Astra Low Emits Zero Reasoning Tokens Yet 2x More Accurate Than Sol Max — rbhar90 · 2026-09-05