EuroEval Catalan NLU: leaderboard
Catalan-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 April 2027. 219 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 68.21 | #66 |
| 2 | Claude Sonnet 4.6 | 67.49 | #90 |
| 3 | Qwen 3 32B (Non-reasoning) | 66.7 | #443 (Qwen 3 32B) |
| 4 | GPT-5.2 | 66.33 | #110 |
| 5 | Gemini 3 Flash (Preview) (Thinking) | 66.23 | #82 (Gemini 3 Flash (Preview)) |
| 6 | GPT-4.1 | 65.77 | #248 |
| 7 | GPT-5.4 Mini (2026-03-17) (High) | 65.64 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 8 | Bielik-11B-v2.3-Instruct | 64.9 | #693 |
| 9 | Qwen 3.5 35B A3B FP8 | 64.75 | #362 |
| 10 | Gemini 2.5 Flash (Thinking) | 64.47 | #245 (Gemini 2.5 Flash) |
| 11 | Qwen 3.5 27B FP8 | 64.47 | #164 |
| 12 | Gemini 2.5 Pro | 64.28 | #152 |
| 13 | Gemini 3 Flash (Preview) (Non-reasoning) | 64.25 | #82 (Gemini 3 Flash (Preview)) |
| 14 | Claude Sonnet 4.5 (Thinking) | 64.23 | #142 (Claude Sonnet 4.5) |
| 15 | Gemini 3.1 Flash Lite (Preview) | 63.98 | #194 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-catalan-nlu · How It Works · Data refreshed daily, snapshot 2026-10-20.