EuroEval Spanish NLU - CoNLL ES: leaderboard
The "CoNLL ES" subtask of EuroEval Spanish NLU. Spanish-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. 412 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 86.12 | #142 (Claude Sonnet 4.5) |
| 2 | GPT-5.4 Mini (2026-03-17) (High) | 83.02 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 3 | Claude 3.7 Sonnet (20250219) (Thinking) | 80.95 | #201 (Claude 3.7 Sonnet (20250219)) |
| 4 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 80.25 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 5 | Gemini 3 Pro (Preview) | 79.88 | #66 |
| 6 | Qwen 3.5 35B A3B Base | 79.68 | #310 |
| 7 | GPT-5 (2025-08-07) (High) | 79.47 | |
| 8 | Qwen 3 4B 2507 (Thinking) | 79.34 | #538 (Qwen 3 4B 2507) |
| 9 | O3 (2025-04-16) | 79.28 | #121 |
| 10 | Claude Sonnet 4.6 | 79.02 | #90 |
| 11 | Gemini 3.1 Flash Lite (Preview) | 78.37 | #194 |
| 12 | GPT-5 | 78.06 | #98 |
| 13 | Qwen 3 14B | 78.03 | #537 |
| 14 | Llama 3.1 70B Instruct | 77.93 | #561 |
| 15 | Qwen 3 30B A3B | 77.85 | #493 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-spanish-nlu-conll-es · How It Works · Data refreshed daily, snapshot 2026-10-20.