EuroEval Serbian Summarization - LR SUM SR — leaderboard
Metric: Score (%). Source: euroeval.com. 148 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Bielik-11B-v2.3-Instruct | 31.95 |
| 2 | Llama 3.1 8B Instruct | 31.53 |
| 3 | Mistral Small 3.1 | 31.09 |
| 4 | Llama 3.3 70B Instruct | 31.08 |
| 5 | Qwen 3 14B (Non-reasoning) | 30.64 |
| 6 | Mistral Small 3.2 | 30.38 |
| 7 | Gemma 3 12B (IT) | 30.34 |
| 8 | Apertus-70B-Instruct-2509 | 30.32 |
| 9 | Qwen 3 30B A3B 2507 Instruct | 30.3 |
| 10 | Qwen 2.5 72B Instruct | 30.27 |
| 11 | GPT-4.1 | 30.11 |
| 12 | Qwen 3 1.7B (Non-reasoning) | 30.1 |
| 13 | Tiny Aya Global | 30.08 |
| 14 | Qwen 3 8B (Non-reasoning) | 30.07 |
| 15 | Gemini 3 Pro (Preview) | 30.06 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-serbian-summarization-lr-sum-sr · How the rankings work · Data refreshed daily, snapshot 2026-07-22.