CodeGolf Bench - Python Pass@1: leaderboard
Metric: Pass@1 (%): share of problems whose Python solution passes every hidden test, estimated from 10 samples per problem; 115 code.golf problems (holes), solutions checked against the platform's hidden tests through its API; golfing prompt, temperature 0.2, top-p 0.95, up to 32,768 output tokens, 10 samples per problem; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek R1 | 43.47 |
| 2 | Llama 4 Maverick | 38.69 |
| 3 | Gemini 2.5 Pro (Preview 03-25) | 38.2 |
| 4 | DeepSeek V3 (0324) | 34.78 |
| 5 | Gemini 2.5 Flash (Preview 04-17) | 34.18 |
| 6 | Qwen 3 235B A22B | 32.94 |
| 7 | Llama 4 Scout | 30.34 |
| 8 | DeepSeek R1 Distill Llama 70B | 29.69 |
| 9 | Gemma 3 27B (IT) | 23.47 |
Interactive version: theaggregate.ai/benchmark?slug=codegolf-bench-python-pass-1 · How It Works · Data refreshed daily, snapshot 2026-10-07.