EuroEval Spanish NLU: leaderboard
Spanish-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: NLU Average Score (%). Source: euroeval.com. Saturation forecast: Around May 2027. 418 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 65.42 | #66 |
| 2 | Claude Sonnet 4.6 | 65.08 | #90 |
| 3 | GPT-5.4 Mini (2026-03-17) (High) | 63.97 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 4 | Gemini 3 Flash (Preview) (Non-reasoning) | 62.13 | #82 (Gemini 3 Flash (Preview)) |
| 5 | Gemini 3 Flash (Preview) (Thinking) | 61.09 | #82 (Gemini 3 Flash (Preview)) |
| 6 | Qwen 3.5 35B A3B Base | 61.08 | #310 |
| 7 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 60.61 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 8 | Qwen 3 4B 2507 (Thinking) | 60.1 | #538 (Qwen 3 4B 2507) |
| 9 | Grok 4.1 Fast (Reasoning) | 60.07 | #217 (Grok 4.1 Fast) |
| 10 | Gemini 3.1 Flash Lite (Preview) | 58.92 | #194 |
| 11 | GPT-5.2 | 58.63 | #110 |
| 12 | Voxtral-Small-24B-2507 | 58.63 | #581 |
| 13 | Mistral Small 3.2 | 58.35 | #605 |
| 14 | Mistral Small 3.1 | 58.31 | #610 |
| 15 | Qwen 3 Next 80B A3B Instruct | 58.04 | #368 |
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 · How It Works · Data refreshed daily, snapshot 2026-10-20.