CodegenBench - LeetSunway: leaderboard

Metric: Pass@1 (%): share of problems whose first generated program compiles, runs and matches the reference output within tolerance; zero-shot through the official APIs with a short optimization system prompt, on the 20 LeetSunway kernels that must use the master-slave core paradigm of the SW26010 processor (Sunway TaihuLight, swgcc); higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.

Top models

#ModelScore
1Claude Opus 4.748
2Claude Opus 4.633
3Qwen 3.6 Plus22
4Claude Sonnet 4.66
5DeepSeek V4 Flash5
6DeepSeek V4 Pro0
7DeepSeek V3.20
8Qwen 3.5 Plus0
9Qwen 3.6 Flash0

Interactive version: theaggregate.ai/benchmark?slug=codegenbench-leetsunway · How It Works · Data refreshed daily, snapshot 2026-09-29.