Unverified50% confidenceFactExact time
Pass@5 is an evaluation metric: generate 5 independent solutions to the same problem, and success is counted if any one of them passes verification; compared with Pass@1, it better reflects the upper limit of capability.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
Unverifiedpass@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66% similarUnverifiedFrontierCode benchmark results from different versions or different sampling strategies (such as pass@1 and pass@k) cannot be directly compared across the board.62% similarUnverifiedIn code generation evaluation, pass@1 measures the probability that the model passes the test with a single generation, while pass@10 allows the model to make 10 attempts and takes the best result62% similarUnverifiedThe pass@k metric is widely adopted on standard benchmarks such as HumanEval and MBPP.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1028132API
curl https://kongchang.com/api/v1/knowledge/claims/1028132MCP
get_claim(id=1028132)