EuroEval English NLU - CoNLL EN: leaderboard
The "CoNLL EN" subtask of EuroEval English NLU. English-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: Estimated already saturated. 333 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash (Preview) (Thinking) | 92.3 | #82 (Gemini 3 Flash (Preview)) |
| 2 | Claude Sonnet 4.5 (Thinking) | 89.97 | #142 (Claude Sonnet 4.5) |
| 3 | Gemini 3 Flash (Preview) (Non-reasoning) | 89.2 | #82 (Gemini 3 Flash (Preview)) |
| 4 | GPT-5 (2025-08-07) (High) | 88.6 | |
| 5 | Claude Sonnet 4.5 (Non-reasoning) | 88.59 | #142 (Claude Sonnet 4.5) |
| 6 | Gemma 4 31B (IT) | 88.53 | #176 |
| 7 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 88.15 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 8 | Gemma 4 26B A4B (IT) | 87.04 | #297 |
| 9 | Gemini 3 Pro (Preview) | 86.89 | #66 |
| 10 | Qwen 3 Next 80B A3B (Thinking) | 86.82 | #320 (Qwen 3 Next 80B A3B) |
| 11 | O3 (2025-04-16) | 86.49 | #121 |
| 12 | Qwen 3 4B 2507 (Thinking) | 86.34 | #538 (Qwen 3 4B 2507) |
| 13 | Qwen 3.5 35B A3B Base | 85.82 | #310 |
| 14 | GPT-5.4 Mini (2026-03-17) (High) | 85.75 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 15 | GPT-5 | 85.59 | #98 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-english-nlu-conll-en · How It Works · Data refreshed daily, snapshot 2026-10-20.