EuroEval Latvian NLU - Fullstack NER LV: leaderboard
The "Fullstack NER LV" subtask of EuroEval Latvian NLU. Latvian-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. 233 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 83.9 | #98 |
| 2 | Gemma 4 31B (IT) | 78.27 | #176 |
| 3 | GPT-5.4 Mini (2026-03-17) (High) | 77.5 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 4 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 77.15 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 5 | Claude Sonnet 4.5 (Thinking) | 76.75 | #142 (Claude Sonnet 4.5) |
| 6 | Qwen 3 4B 2507 (Thinking) | 73.2 | #538 (Qwen 3 4B 2507) |
| 7 | GLM 4.5 Air (Thinking) | 71.41 | #392 (GLM 4.5 Air) |
| 8 | Claude Sonnet 4.5 (Non-reasoning) | 71.39 | #142 (Claude Sonnet 4.5) |
| 9 | Gemma 4 26B A4B (IT) | 69.43 | #297 |
| 10 | Claude Sonnet 4.6 | 67.56 | #90 |
| 11 | Qwen 3 Next 80B A3B (Thinking) | 67.52 | #320 (Qwen 3 Next 80B A3B) |
| 12 | Llama 3.1 70B | 66.65 | #592 |
| 13 | Qwen 3 30B A3B | 64.85 | #493 |
| 14 | Gemini 3 Pro (Preview) | 64.6 | #66 |
| 15 | Gemini 3.1 Flash Lite (Preview) | 64.3 | #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-latvian-nlu-fullstack-ner-lv · How It Works · Data refreshed daily, snapshot 2026-10-20.