EuroEval Czech Knowledge: leaderboard
EuroEval's Czech knowledge task — multiple-choice general-knowledge questions in Czech, from the EuroEval European language model benchmark suite.
Metric: Knowledge Average Score (%). Source: euroeval.com. Saturation forecast: Around March 2027. 211 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 100 | #152 |
| 2 | GPT-5 | 100 | #98 |
| 3 | Gemini 3 Pro (Preview) | 100 | #66 |
| 4 | GPT-5 Mini (2025-08-07) | 100 | #172 |
| 5 | Claude Sonnet 4.5 (Thinking) | 100 | #142 (Claude Sonnet 4.5) |
| 6 | Gemini 2.5 Flash (Thinking) | 100 | #245 (Gemini 2.5 Flash) |
| 7 | GPT-5.4 Mini (2026-03-17) (High) | 100 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 8 | Gemini 3 Flash (Preview) (Thinking) | 100 | #82 (Gemini 3 Flash (Preview)) |
| 9 | Grok 4.20 Beta (0309) (Reasoning) | 100 | #204 (Grok 4.20 Beta (0309)) |
| 10 | GPT-5.4 Nano (2026-03-17) (High) | 94.63 | #266 (GPT-5.4 Nano (2026-03-17)) |
| 11 | Claude Sonnet 4.6 | 94.42 | #90 |
| 12 | Gemini 3.1 Flash Lite (Preview) | 94.42 | #194 |
| 13 | Gemini 3 Flash (Preview) (Non-reasoning) | 93.93 | #82 (Gemini 3 Flash (Preview)) |
| 14 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 93.88 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 15 | GPT-5 Nano | 93.75 | #425 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-czech-knowledge · How It Works · Data refreshed daily, snapshot 2026-10-20.