AItonomy Foundation 发布 Science's Last Exam（SLE）基准，用真实研究目标和实验室数据、软件等工具评估智能体的端到端科研能力，并由隐藏的专家检查评分。Aran Komatsuzaki 转述称，该基准包含 9 个学科的 91 个任务，Claude Opus 5 得 67.0、GPT-6.1 Sol 得 64.25、Kimi K3.1 Pro 得 58.25，评估范围覆盖端到端实验方案规划、实验观察、科学计算和软件操作。他认为这是对前沿 LLM 评测的有益补充，但 SOTA 模型已达 60 分以上，需要更难的题目。
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.
【引用 @aitonomyfound】I’m incredibly proud to share what we’ve been building at the AItonomy Foundation: Science’s Last Exam (SLE). 🧪✨
Can AI do science? Instead of Q&A, we give agents actual research goals and real-world tools (lab data, software), then grade them against hidden expert checks.
Deeply grateful to the brilliant minds at 34 institutions (Harvard, MIT, Stanford, etc.) who helped craft 113 tasks across 9 disciplines. ❤️
The results? Science is hard. Of 8 frontier models, the best scored 37.5/100. On true end-to-end research? All scored 0. We have an exciting journey ahead! 🚀
(Claude Opus 5.5 & GPT-6 Astra coming soon—just clearing bio safeguards).
Everything opens Nov 6. 🗓️
🌐 https://sciences-last-exam.org
💻 https://github.com/Gen-Verse/Science-Last-Exam
🤗 https://huggingface.co/datasets/Gen-Verse/Science-Last-Exam
Thread below on how the exam works and where models break. 👇 (1/14)
#AGI #AIforScience #AIAgents #SciencesLastExam #OpenAI #Anthropic
I’m incredibly proud to share what we’ve been building at the AItonomy Foundation: Science’s Last Exam (SLE). 🧪✨ Can AI do science? Instead of Q&A, we give agents actual research goals and real-world tools (lab data, software), then grade them against hidden expert checks. Deeply grateful to the brilliant minds at 34 institutions (Harvard, MIT, Stanford, etc.) who helped craft 113 tasks across 9 disciplines. ❤️ The results? Science is hard. Of 8 frontier models, the best scored 37.5/100. On true end-to-end research? All scored 0. We have an exciting journey ahead! 🚀 (Claude Opus 5.5 & GPT-6 Astra coming soon—just clearing bio safeguards). Everything opens Nov 6. 🗓️ 🌐 https://sciences-last-exam.org 💻 https://github.com/Gen-Verse/Science-Last-Exam 🤗 https://huggingface.co/datasets/Gen-Verse/Science-Last-Exam Thread below on how the exam works and where models break. 👇 (1/14) #AGI #AIforScience #AIAgents #SciencesLastExam #OpenAI #Anthropic在 X 查看被引用的帖子
来源：Aran Komatsuzaki · x.com