GPT-6 Astra scores 66% on ARC-AGI-3, nears 100% with persistent conversation

François Chollet 评 GPT-6 Astra 在 ARC-AGI-3 上的表现

François Chollet reports GPT-6 Astra's ARC-AGI-3 results. Standard harness yields 66%; persistent conversation with custom compaction pushes it near 100%. Cost is roughly $360 per task. The post doesn't disclose task count or latency. Worth noting: near-perfect scores rely on a conversational harness, not raw model output, so it's not yet general reasoning out of the box.

Why it matters: Chollet himself posted GPT-6 Astra's ARC-AGI-3 scores: 66% standard, near-100% with multi-turn dialogue and custom context compression, at ~$360 per task. The contrast and the cost make it a must-read for the reasoning-eval crowd, but it's not single-pass reasoning, so it does...