Unverified50% confidenceFactExact time
The pass@k metric is widely adopted on standard benchmarks such as HumanEval and MBPP.
1
Sources
50%
Confidence
Long-term
Relevance
9/22/2026
First Seen
Sources
Related Entities
Related Claims
Unverified当前主流基准测试体系如HumanEval、MBPP等代码评估集往往关注平均性能，对分布式的场景特定退化现象敏感度不足70% similarUnverifiedGLM-5.2 performs prominently on mainstream coding benchmarks such as HumanEval and MBPP.67% similarUnverifiedpass@k指标衡量模型生成k个候选代码片段时至少有一个能通过全部测试用例的概率65% similarUnverifiedTo evaluate whether a local model is suitable for your needs, you can refer to standardized benchmark scores such as MMLU and HumanEval.64% similarUnverifiedpass@1指标最初从代码生成评测领域（如HumanEval基准）借鉴而来，用于衡量模型一次生成正确代码的概率64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/940039API
curl https://kongchang.com/api/v1/knowledge/claims/940039MCP
get_claim(id=940039)