AI 导读
ARC Prize 公布 GPT-6.1 Sol 在 ARC-AGI (Verified) 上的成绩:ARC-AGI-3 为 52.7%/$7.6K(标准 harness)、96.4%/$4.4K(provider adapter harness),ARC-AGI-2 为 94.2%/$0.25/task,ARC-AGI-1 为 98.5%/$0.06/task。其 ARC-AGI-3 的 96.4% 与 GPT-6 Astra 的 99.9% 相当,但成本低 77%。
正文
GPT-6.1 Sol from @OpenAI on ARC-AGI (Verified):
- ARC-AGI-3: 52.7%, $7.6K (standard harness), 96.4%, $4.4K (provider adapter harness)
- ARC-AGI-2: 94.2%, $0.25/task
- ARC-AGI-1: 98.5%, $0.06/task
Its 96.4% on v3 was comparable to GPT-6 Astra's 99.9% but at a 77% lower cost.
来源:ARC Prize · x.com