CL4SE - Code Summarization: leaderboard
Metric: ROUGE-L F1 (%) of the generated docstring against the original one for 8,225 Python functions from the 10 most-starred Python repositories; zero-shot; temperature 0; CL4SE (context learning for software engineering); higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 Max | 22.75 | #201 |
| 2 | Claude 3.5 Haiku (20241022) | 20.86 | #574 |
| 3 | GPT-OSS-120B | 18.96 | #330 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=cl4se-code-summarization · How It Works · Data refreshed daily, snapshot 2026-10-11.