French LLM Leaderboard - BAC FR: leaderboard
French BAC-style exam score, measuring French-language school exam reasoning and answer accuracy.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 55 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek R1 Distill Llama 70B | 50.71 | #594 |
| 2 | Mistral Large 2 (Nov) Instruct (2411) | 50.14 | #465 |
| 3 | DeepSeek R1 Distill Qwen 32B | 45.62 | #655 |
| 4 | Llama 3.3 70B Instruct | 45.34 | #534 |
| 5 | Phi-4 | 45.34 | #706 |
| 6 | Qwen 2.5 72B Instruct | 44.92 | #454 |
| 7 | Qwen 2.5 14B Instruct | 43.64 | #647 |
| 8 | Mistral Small 3 | 43.08 | #627 |
| 9 | Mixtral 8x22B Instruct (v0.1) | 42.23 | #756 |
| 10 | calme-3.2-instruct-78B | 41.81 | #347 |
| 11 | Falcon3-10B-Instruct | 39.69 | #1157 |
| 12 | Phi-3.5-MoE-instruct | 36.3 | #700 |
| 13 | Falcon3-7B-Instruct | 35.31 | #1305 |
| 14 | Phi-3-medium-128k-instruct | 33.76 | #1013 |
| 15 | granite-3.1-8B-instruct | 32.77 | #1069 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=french-llm-leaderboard-bac-fr · How It Works · Data refreshed daily, snapshot 2026-10-20.