EuroEval French NLU - FQuAD: leaderboard
The "FQuAD" subtask of EuroEval French NLU. French-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Character F1 (%). Source: euroeval.com. Saturation forecast: Not forecast. 374 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.1 405B | 77.36 | #466 |
| 2 | Llama 3.1 70B | 76.72 | #592 |
| 3 | OLMo-2-1124-7B | 76.56 | #1401 |
| 4 | Llama 3 70B | 76.09 | #631 |
| 5 | gemma-3-27B-pt | 76.08 | #502 |
| 6 | Qwen 3 32B (Non-reasoning) | 75.96 | #443 (Qwen 3 32B) |
| 7 | Gemma 2 27B | 75.88 | #806 |
| 8 | Yi 1.5 34B | 75.71 | #807 |
| 9 | Qwen 3 14B (Non-reasoning) | 75.59 | #537 (Qwen 3 14B) |
| 10 | Bielik-11B-v2.3-Instruct | 75.46 | #693 |
| 11 | Llama 2 70B Base | 75.37 | #1025 |
| 12 | Mistral Small 3.1 | 75.02 | #610 |
| 13 | aya-23-35B | 74.54 | #883 |
| 14 | Gemma 3n E4B | 74.51 | #878 |
| 15 | Mixtral 8x7B (v0.1) | 74.48 | #1083 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-french-nlu-fquad · How It Works · Data refreshed daily, snapshot 2026-10-20.