CodegenBench - BLAS Kunpeng: 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 106 BLAS subroutines on the ARM-based Kunpeng processor (SVE, SME and on-die HBM; bisheng compiler, kblas); higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.

Top models

#ModelScore
1Claude Opus 4.776
2Claude Opus 4.667
3Claude Sonnet 4.661
4Qwen 3.6 Plus22
5DeepSeek V4 Pro21
6DeepSeek V3.213
7DeepSeek V4 Flash11
8Qwen 3.6 Flash9
9Qwen 3.5 Plus6

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