Expired50% confidenceBenchmarkExact time
On professional code benchmarks such as HumanEval, the performance of top models has approached or even surpassed average human-level performance.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/18/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/16/2026(expired)
Sources
Related Claims
UnverifiedDomestic models such as GLM5.1 have achieved scores on standard programming benchmarks like HumanEval and MBPP that are close to or even surpass some international competitors.72% similarVerifiedPass rates of code generation models on benchmarks such as HumanEval jumped from 60% to over 90% between 2024 and 202569% similarUnverified模型在标准化评测基准（如HumanEval、MMLU）上的高分表现与真实工作场景中的可靠性之间存在显著差距68% similarUnverifiedBenchmark overfitting is a core problem in current large model evaluation; some models achieve pass rates above 90% on HumanEval but perform far below expectations in real codebases.68% similarUnverifiedWork sample tests have been shown in industrial-organizational psychology to have the highest predictive validity, with a correlation coefficient of r=0.5466% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/549438API
curl https://kongchang.com/api/v1/knowledge/claims/549438MCP
get_claim(id=549438)