Beta version: *Information might not be fully accurate. Please report any discrepancies.

HumanEval

human-evalCodingHigher is better

Functional correctness of synthesized programs from docstrings.

Loading chart…

Loading leaderboard…

Benchmark Info

Current SOTA

Best Open Source

Median Score88.1%

DistributionClustered

Score Distribution

P10

P50

P90

0100%

SWE-bench Verified

swe-bench-verified

BigCodeBench

bigcodebench

Codeforces

codeforces

LiveCodeBench v6

livecodebench-v6

LiveCodeBench Pro

livecodebench-pro

AA Coding Index

aa-coding-index