EuroEval Estonian NLU - EstNER: leaderboard
The "EstNER" subtask of EuroEval Estonian NLU. Estonian-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 September 2027. 250 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 4 31B (IT) | 58.07 | #176 |
| 2 | GPT-5.4 Mini (2026-03-17) (High) | 57.99 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 3 | Voxtral-Small-24B-2507 | 57.89 | #581 |
| 4 | gemma-3-27B-pt | 57.31 | #502 |
| 5 | Llama 3.1 70B | 57.15 | #592 |
| 6 | Claude Sonnet 4.5 (Thinking) | 56.74 | #142 (Claude Sonnet 4.5) |
| 7 | Magistral Small 1.2 | 56.59 | #570 |
| 8 | GPT-5 | 56.4 | #98 |
| 9 | Mistral Small 3.2 | 56.26 | #605 |
| 10 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 56.17 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 11 | Llama 3.1 70B Instruct | 55.96 | #561 |
| 12 | Gemini 3 Flash (Preview) (Thinking) | 55.72 | #82 (Gemini 3 Flash (Preview)) |
| 13 | Llama 3.3 70B Instruct | 55.71 | #534 |
| 14 | GPT-5 Nano | 55.07 | #425 |
| 15 | Mistral Small 3.1 | 54.69 | #610 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-estonian-nlu-estner · How It Works · Data refreshed daily, snapshot 2026-10-20.