GPT 6 Astra and Fable 5.1 got 20 attempts to pick up a block and place it into a bowl

GPT 6 Astra and Fable 5.1 got 20 attempts to pick up a block and place it into a bowl

Astra: 95% success, 2.5 min, $0.94 per run

Fable 5.1: 40% success, 6.8 min, $2.12 per run

Astra also used 6× fewer output tokens, just 2.1k per run.