Batched moves let GPT-6 plan 10-20 steps ahead, cutting 3B tokens to 350M
patience_cave · x · 2026-09-07
MazeBench allowed GPT-6 Astra to submit moves in batches instead of one per turn. The model consistently planned 10-20 moves ahead, collapsing 3 billion tokens into 350M. On the Klotski puzzle, batched Astra solved it 3x cheaper but took 60% longer to think.
More from Models
- Google's Astra Agent Allegedly Crushes Existing CAPTCHAs, Sparking Rethink of Bot Checks — eyishazyer · 2026-09-07
- Intelligence price collapsed ~1000x in 18 months as models keep getting smarter — ccerrato147 · 2026-09-07
- Blogger says Google Astra's natural tone broke his last dependency on Claude — StewartalsopIII · 2026-09-07
- VC slams frontier lab subscriptions: usage limits to worsen, top models to shrink — StewartalsopIII · 2026-09-07
- GPT Astra scores just 13% on MazeBench without tools — Wonderful_Buffalo_32 · 2026-09-07
- GPT-6 Astra Generates Stunning Math Animation in One Shot — omarsar0 · 2026-09-07