EuroEval Finnish Summarization - Xlsum FI: leaderboard
EuroEval's Finnish summarization task on the Xlsum FI dataset — models write abstractive summaries of Finnish news articles, scored with automatic text-overlap metrics, from the EuroEval European language model benchmark suite.
Metric: chrF3++ (%). Source: euroeval.com. Saturation forecast: Around 2035. 151 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash (Preview) (Non-reasoning) | 32.86 | #82 (Gemini 3 Flash (Preview)) |
| 2 | Gemini 3 Pro (Preview) | 32.77 | #66 |
| 3 | Gemini 3 Flash (Preview) (Thinking) | 32.73 | #82 (Gemini 3 Flash (Preview)) |
| 4 | Gemini 3.1 Flash Lite (Preview) | 32.72 | #194 |
| 5 | Gemma 3 27B (IT) | 32.55 | #523 |
| 6 | Gemma 4 26B A4B (IT) | 32.22 | #297 |
| 7 | GPT-4.1 | 32.16 | #248 |
| 8 | Gemma 4 31B (IT) | 32.16 | #176 |
| 9 | Gemma 3 12B (IT) | 32.09 | #664 |
| 10 | Claude Sonnet 4.6 | 31.94 | #90 |
| 11 | Claude Sonnet 4.5 (Non-reasoning) | 31.85 | #142 (Claude Sonnet 4.5) |
| 12 | Qwen 3.5 35B A3B Base | 31.78 | #310 |
| 13 | Claude Sonnet 4.5 (Thinking) | 31.66 | #142 (Claude Sonnet 4.5) |
| 14 | Ministral-3-14B-Instruct-2512 | 31.36 | #615 |
| 15 | Qwen 3.5 35B A3B FP8 | 31.15 | #362 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-finnish-summarization-xlsum-fi · How It Works · Data refreshed daily, snapshot 2026-10-20.