Unverified50% confidenceBenchmarkExact time
Claude 5.1在Humanities Last Exam不带工具时得分60.9%，接入工具后升至65%
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/4/2026
First Seen
Valid until: 12/3/2026
Sources
Related Entities
Related Claims
Unverified在Terminal-Bench 4.0上，Claude 5.1标准版得分55.8%，受信版达到60.9%；OS World 2.0部分完成率为77.9%66% similarUnverified在长周期智能体工作流测试（Engines Last Exam）中，GPT-5.6 Sol 得分 53.6，比 Faber-5 高出 13.1 分66% similarUnverified在Agents' Last Exam评测中，GPT-5.6 Sol取得53.6分，比采用自适应推理的Claude Fable 5高出13.1分64% similarUnverifiedClaude 5.1在Terminal-Bench Science上得分52.6%，相比上一代的24.7%实现翻倍以上跃升64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/853482API
curl https://kongchang.com/api/v1/knowledge/claims/853482MCP
get_claim(id=853482)