gpt-6IMAGE
Astra 刷满 ARC AGI 3 后，有人提议上千年数学难题当新基准
作者调侃 GPT-6 Astra 已饱和 ARC AGI 3，于是拿克雷数学研究所的千禧年难题当新基准，称 Astra 与 Anthropic Fable 都只做到约 4%。
samagra14
@samagra_sharma2026.09.14收录原推发布于 2026.09.04
samagra14 @samagra_sharma原推发布于 2026.09.04
Introducing ARC AGI 4 Now that GPT-6 Astra has already saturated the Arc AGI 3 benchmarks, it is time we introduce a newer more foundational benchmark to measure frontier intelligence. @AnthropicAI Fable and @ChatGPT Astra are both tied at ~4%. https://www.claymath.org/millennium-problems/
x.com/samagra_sharma/status/2095743467435761886采集于 2026.09.14 15:30