Unverified50% confidenceOpinionExact time
HumanEval、SWE-bench等代码评测基准的排名是评估执行型Agent底层模型选型的重要参考维度
1
Sources
50%
Confidence
Long-term
Relevance
7/6/2026
First Seen
Sources
Connecting an Open-Source AI Assistant to OpenClaw: A Complete Local Agent Setup Guide
bilibili枫影剑wind7/4/2026
Related Claims
UnverifiedHumanEval, MBPP, and SWE-bench are standard benchmark datasets specifically targeting code generation.79% similarUnverifiedHumanEval, MBPP, and SWE-bench are industry-standard benchmarks in the field of AI programming evaluation.76% similarUnverifiedModel evaluations in the code generation field, such as HumanEval and SWE-bench, place special emphasis on functional correctness rather than fluency.76% similarVerifiedMainstream benchmarks for evaluating programming large models include HumanEval, MBPP, SWE-bench, and LiveCodeBench76% similarUnverified锚定样本技术在评估提示词中加入已由人类专家标注明确分数的示例，覆盖评分量表各个档位，可帮助裁判模型准确理解评分标准76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/114299API
curl https://kongchang.com/api/v1/knowledge/claims/114299MCP
get_claim(id=114299)