Unverified50% confidenceFactExact time
SWE-bench and LiveCodeBench are relatively well-recognized industry benchmarks for evaluating code large language models
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedMainstream benchmarks for evaluating programming large models include HumanEval, MBPP, SWE-bench, and LiveCodeBench79% similarUnverifiedLiveCodeBench continuously crawls new problems from competitive programming platforms such as LeetCode, Codeforces, and AtCoder to prevent models from gaming the benchmark, and its evaluation covers four dimensions: code generation, self-repair, code execution reasoning, and test output prediction.74% similarUnverifiedSWE-bench is considered by the industry to be the most practically valuable code benchmark due to its closeness to real engineering scenarios.72% similarUnverifiedHumanEval, MBPP, and SWE-bench are standard benchmark datasets specifically targeting code generation.69% similarUnverifiedSWE-bench相比HumanEval、MBPP等传统基准更贴近实际开发场景，考验模型对大型代码库的理解、跨文件依赖追踪和缺陷定位能力67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/910355API
curl https://kongchang.com/api/v1/knowledge/claims/910355MCP
get_claim(id=910355)