Kurakurai French LLM - MBPP: leaderboard
Evaluates small French-language models on the Kurakurai French LLM leaderboard, a static board of French translations and French subsets of standard benchmarks, filtered by size class up to about 4B parameters. Scores are percentages averaged over ten sampling seeds, decoded at temperature 0.6, top_p 0.95, top_k 20, with thinking disabled and 16k context. This series is the board's MBPP column, which upstream documents as MBPP+ (French) code generation. These are French scores, comparable to each other but not to published English numbers for the same models. This series reports Kurakurai French LLM - MBPP using pass@1 (%).
Metric: pass@1 (%). Source: huggingface.co. Saturation forecast: Around February 2027. 42 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 4B 2507 Instruct | 76.03 | #768 |
| 2 | Gemma 4 E2B (IT) | 68.68 | #779 |
| 3 | Granite 4.1 3B | 68.07 | #899 |
| 4 | Qwen 3.5 4B | 67.75 | #514 |
| 5 | granite-4.0-h-micro | 66.32 | #1078 |
| 6 | Gemma 3 4B (IT) | 62.04 | #997 |
| 7 | Qwen 3 1.7B | 57.33 | #1175 |
| 8 | SmolLM3-3B | 56.14 | #1239 |
| 9 | LFM2 2.6B | 54.42 | #1313 |
| 10 | LFM2.5-1.2B Instruct | 48.65 | #1451 |
| 11 | Llama 3.2 3B Instruct | 44.26 | #1327 |
| 12 | Gemma 3 1B (IT) | 43.86 | #1481 |
| 13 | Qwen 3.5 2B | 42.12 | #1114 |
| 14 | Phi-4 Mini Instruct | 39.18 | #1232 |
| 15 | Qwen 3 0.6B | 37.41 | #1448 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=kurakurai-french-llm-mbpp · How It Works · Data refreshed daily, snapshot 2026-11-05.