CodeGolf Bench: leaderboard
This paper introduces Code Bench, a benchmark capable of evaluating Large Language Models (LLMs) concise code generation abilities in 60 programming languages. Based on code golf, a recreational programming competition focused on minimal character or byte solutions, the benchmark provides a distinctive measure of LLMs ability to produce efficient, concise code. Unlike existing benchmarks limited b
Metric: C++ Best %ile (self-reported). Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 03-25) | 73.9 | #123 |
| 2 | DeepSeek R1 | 60.21 | #251 |
| 3 | Gemini 2.5 Flash (Preview 04-17) | 55.19 | #237 |
| 4 | Qwen 3 235B A22B | 48.24 | #313 |
| 5 | Llama 4 Maverick | 44.85 | #464 |
| 6 | DeepSeek R1 Distill Llama 70B | 44.18 | #594 |
| 7 | DeepSeek V3 (0324) | 38 | #340 |
| 8 | Gemma 3 27B (IT) | 20.92 | #523 |
| 9 | Llama 4 Scout | 16.65 | #656 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=codegolf-bench · How It Works · Data refreshed daily, snapshot 2026-10-20.