OpenAI发布GPT-6.1 Sol模型
精选理由
OpenAI新模型GPT-6.1 Sol在ARC-AGI测试中表现优异，成本比前代低77%，性价比突出。
OpenAI发布了GPT-6.1 Sol模型，在ARC-AGI-3基准测试中达到52.7%准确率，获得7600美元奖金。在标准测试套件中得分为96.4%，成本比GPT-6 Astra低77%。在ARC-AGI-2测试中得分为94.2%，每任务成本0.25美元。在ARC-AGI-1测试中得分为98.5%，每任务成本仅0.06美元。
原文 · ARC Prize
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. https://t.co/GwZbhX5oxd
