EuroEval Bosnian NLU - MMS BS: leaderboard
The "MMS BS" subtask of EuroEval Bosnian NLU. Bosnian-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Matthews correlation coefficient (%). Source: euroeval.com. Saturation forecast: Around March 2027. 212 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 Mini | 56.43 | #355 |
| 2 | Gemma 4 31B (IT) | 53.93 | #176 |
| 3 | Gemini 2.5 Flash (Thinking) | 53.82 | #245 (Gemini 2.5 Flash) |
| 4 | GPT-5 Nano | 53.18 | #425 |
| 5 | GPT-5.4 Nano (2026-03-17) (High) | 52.99 | #266 (GPT-5.4 Nano (2026-03-17)) |
| 6 | Gemini 2.5 Flash Lite (Non-reasoning) | 52.69 | #422 (Gemini 2.5 Flash Lite) |
| 7 | GPT-5 Mini (2025-08-07) | 52.39 | #172 |
| 8 | Gemini 3 Flash (Preview) (Thinking) | 52.05 | #82 (Gemini 3 Flash (Preview)) |
| 9 | Gemini 3 Pro (Preview) | 51.37 | #66 |
| 10 | Qwen 3.5 27B FP8 | 51.36 | #164 |
| 11 | GPT-5.2 | 51.12 | #110 |
| 12 | GPT-5 | 50.99 | #98 |
| 13 | Gemini 3.1 Pro (Preview) | 50.52 | #58 |
| 14 | Gemini 2.5 Pro | 50.26 | #152 |
| 15 | Grok 4.1 Fast (Reasoning) | 50.2 | #217 (Grok 4.1 Fast) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-bosnian-nlu-mms-bs · How It Works · Data refreshed daily, snapshot 2026-10-20.