EuroEval French Summarization - Orange SUM: leaderboard
Metric: Score (%). Source: euroeval.com. 151 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 39.53 |
| 2 | Claude Sonnet 4.5 (Non-reasoning) | 39.5 |
| 3 | Gemma 4 26B A4B (IT) | 39.1 |
| 4 | Gemma 4 31B (IT) | 38.97 |
| 5 | GPT-4.1 | 38.87 |
| 6 | Gemini 3.1 Flash Lite (Preview) | 38.63 |
| 7 | Mistral Small 3.1 | 38.55 |
| 8 | Llama 3.3 70B Instruct | 38.39 |
| 9 | Claude Haiku 4.5 (20251001) | 38.37 |
| 10 | Claude Sonnet 4.5 (Thinking) | 38.28 |
| 11 | Grok 4.1 Fast (Reasoning) | 38.23 |
| 12 | gemma-4-E4B-it | 38.19 |
| 13 | Apertus-70B-Instruct-2509 | 38.02 |
| 14 | Qwen 2.5 72B Instruct | 37.98 |
| 15 | Ministral-3-14B-Instruct-2512 | 37.98 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-french-summarization-orange-sum · How It Works · Data refreshed daily, snapshot 2026-09-05.