EuroEval Greek Summarization - Greek Wikipedia: leaderboard
EuroEval's Greek summarization task on the Greek Wikipedia dataset — models write abstractive summaries of Greek 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. 147 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.1 70B | 36.77 | #592 |
| 2 | Qwen 3.5 35B A3B Base | 33.33 | #310 |
| 3 | Gemini 3.1 Flash Lite (Preview) | 32.85 | #194 |
| 4 | Gemini 2.5 Flash Lite (Thinking) | 32.59 | #422 (Gemini 2.5 Flash Lite) |
| 5 | Llama 3.3 70B Instruct | 32.54 | #534 |
| 6 | Claude Haiku 4.5 (20251001) | 32.52 | #261 |
| 7 | GPT-4.1 | 32.49 | #248 |
| 8 | Ministral-3-14B-Instruct-2512 | 32.49 | #615 |
| 9 | Gemini 3 Flash (Preview) (Thinking) | 32.37 | #82 (Gemini 3 Flash (Preview)) |
| 10 | Ministral-3-8B-Instruct-2512 | 32.23 | #661 |
| 11 | Claude Sonnet 4.5 (Thinking) | 32.17 | #142 (Claude Sonnet 4.5) |
| 12 | EuroLLM-22B-Instruct-2512 | 32.15 | #701 |
| 13 | Gemini 3 Pro (Preview) | 31.97 | #66 |
| 14 | EuroLLM-9B-Instruct | 31.92 | #1054 |
| 15 | GPT-5 Mini (2025-08-07) | 31.77 | #172 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-greek-summarization-greek-wikipedia · How It Works · Data refreshed daily, snapshot 2026-10-20.