EuroEval French NLU - Eltec: leaderboard
The "Eltec" subtask of EuroEval French NLU. French-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Micro F1 excluding MISC (%). Source: euroeval.com. Saturation forecast: Around January 2027. 375 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 (Non-reasoning) | 80.67 | #142 (Claude Sonnet 4.5) |
| 2 | Claude Sonnet 4.5 (Thinking) | 79.18 | #142 (Claude Sonnet 4.5) |
| 3 | Gemma 4 31B (IT) | 77.35 | #176 |
| 4 | GPT-5.4 Mini (2026-03-17) (High) | 76.97 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 5 | OLMo 3 7B (Thinking) | 76.8 | #1112 (OLMo 3 7B) |
| 6 | Gemini 3 Pro (Preview) | 76.68 | #66 |
| 7 | Qwen 3.5 35B A3B Base | 75.51 | #310 |
| 8 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 75.42 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 9 | Qwen 3 Next 80B A3B (Thinking) | 75.31 | #320 (Qwen 3 Next 80B A3B) |
| 10 | Gemma 4 26B A4B (IT) | 73.59 | #297 |
| 11 | Qwen 3 4B 2507 (Thinking) | 73.38 | #538 (Qwen 3 4B 2507) |
| 12 | NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 | 73.31 | #679 |
| 13 | Grok 4.1 Fast (Reasoning) | 73.16 | #217 (Grok 4.1 Fast) |
| 14 | Claude Sonnet 4.6 | 73.13 | #90 |
| 15 | GPT-5 Nano | 72.81 | #425 |
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-eltec · How It Works · Data refreshed daily, snapshot 2026-10-20.