EuroEval Serbian Summarization - LR SUM SR: leaderboard

Metric: Score (%). Source: euroeval.com. 148 models tracked.

Top models

#ModelScore
1Llama 3.1 8B Instruct31.53
2Mistral Small 3.131.09
3Llama 3.3 70B Instruct31.08
4Mistral Small 3.230.38
5Gemma 3 12B (IT)30.34
6Apertus-70B-Instruct-250930.32
7Qwen 3 30B A3B 2507 Instruct30.3
8Qwen 2.5 72B Instruct30.27
9GPT-4.130.11
10Tiny Aya Global30.08
11Gemini 3 Pro (Preview)30.06
12granite-4.0-h-micro30.02
13Gemma 3 27B (IT)29.94
14Claude Sonnet 4.5 (Non-reasoning)29.79
15Gemma 3n E4B (IT)29.66

Interactive version: theaggregate.ai/benchmark?slug=euroeval-serbian-summarization-lr-sum-sr · How It Works · Data refreshed daily, snapshot 2026-09-05.