EuroEval Bosnian Summarization - LR SUM BS: leaderboard
EuroEval's Bosnian summarization task on the LR SUM BS dataset — models write abstractive summaries of Bosnian news articles, scored with automatic text-overlap metrics, from the EuroEval European language model benchmark suite.
Metric: chrF3++ (%). Source: euroeval.com. Saturation forecast: Not forecast. 153 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Bielik-11B-v2.3-Instruct | 32.58 | #693 |
| 2 | occiglot-7B-eu5-instruct | 32.23 | |
| 3 | Llama 3.1 8B Instruct | 32.19 | #1035 |
| 4 | Mistral Small 3.1 | 31.94 | #610 |
| 5 | Gemma 3 270M (IT) | 31.58 | #1581 |
| 6 | Voxtral-Small-24B-2507 | 31.36 | #581 |
| 7 | EuroLLM-9B-Instruct | 31.2 | #1054 |
| 8 | Qwen 2.5 72B Instruct | 31.16 | #454 |
| 9 | Mistral Small 3.2 | 31.09 | #605 |
| 10 | Apertus-70B-Instruct-2509 | 31.06 | #820 |
| 11 | Qwen 3 14B (Non-reasoning) | 30.93 | #537 (Qwen 3 14B) |
| 12 | Qwen 3 4B (Non-reasoning) | 30.91 | #829 (Qwen 3 4B) |
| 13 | Qwen 3 1.7B (Non-reasoning) | 30.79 | #1196 (Qwen 3 1.7B) |
| 14 | Qwen 3 8B (Non-reasoning) | 30.6 | #674 (Qwen 3 8B) |
| 15 | cogito-v1-preview-qwen-14B | 30.5 |
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-summarization-lr-sum-bs · How It Works · Data refreshed daily, snapshot 2026-10-20.