CUDAHercules - Module-Level Mean Speed (One-Shot): leaderboard

Metric: Mean expert-relative speed (the expert reference runtime divided by the generated code's runtime, averaged over the 43 architecture-general Class 2 module-level tasks, a task whose code fails the validator counting 0, so 1 matches the expert; one-shot: task specification, hardware metadata and few-shot exemplars, one sample, no execution feedback). Source: arxiv.org. Saturation forecast: Around July 2027. 4 models tracked.

Top models

#ModelScore
1GPT-5.40.43
2Claude Opus 4.60.28
3Qwen 3.5 122B A10B0.01

Interactive version: theaggregate.ai/benchmark?slug=cudahercules-module-level-mean-speed-one-shot · How It Works · Data refreshed daily, snapshot 2026-09-29.