Models

Science's Last Exam 基准发布:91 项任务测端到端科研能力

Why we picked it

新基准 SLE 用 91 个真实科研任务测模型,Claude Opus 5 以 67 分领先 GPT-6.1 Sol 和 Kimi K3.1 Pro,做科研自动化评估的可以看看。

AItonomy Foundation 发布 Science's Last Exam(SLE),一个评估自主端到端科学研究的基准,包含 9 个学科的 91 项任务。基准考察协议规划、实验观察、科学计算和软件操作四个环节。Claude Opus 5 得分 67.0,GPT-6.1 Sol 得分 64.25,Kimi K3.1 Pro 得分 58.25。发布者指出 SotA 模型已在该基准上超过 60 分,后续需要更难的任务。

Original · Aran Komatsuzaki (论文推介)

AItonomy Foundation presents Science's Last Exam (SLE), a benchmark for autonomous end-to-end scientific research.

- 91 tasks across 9 disciplines - Claude Opus 5: 67.0, GPT-6.1 Sol: 64.25, Kimi K3.1 Pro: 58.25 - Evaluates end-to-end protocol planning, experiment observation, scientific computing, and software operation

Great addition to frontier LLM evaluation. Very relevant to RSI/ASI. One caveat is that SotA models already achieve 60+ on this bench, so we need tougher tasks soon.