EuroEval Polish Knowledge: leaderboard

EuroEval's Polish knowledge task — multiple-choice general-knowledge questions in Polish, from the EuroEval European language model benchmark suite.

Metric: Knowledge Average Score (%). Source: euroeval.com. Saturation forecast: Around March 2027. 216 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro (Preview)89.17#66
2Gemini 3 Flash (Preview) (Thinking)87#82 (Gemini 3 Flash (Preview))
3GPT-586.45#98
4Gemini 2.5 Flash (Thinking)85.67#245 (Gemini 2.5 Flash)
5Gemini 2.5 Pro84.99#152
6Gemini 3 Flash (Preview) (Non-reasoning)84.28#82 (Gemini 3 Flash (Preview))
7GPT-5.4 Mini (2026-03-17) (High)83.73#169 (GPT-5.4 Mini (2026-03-17))
8Claude Sonnet 4.5 (Thinking)82.05#142 (Claude Sonnet 4.5)
9Claude Sonnet 4.681.99#90
10Gemini 3.1 Flash Lite (Preview)81.19#194
11Grok 4.1 Fast (Reasoning)80.05#217 (Grok 4.1 Fast)
12GPT-5.4 Nano (2026-03-17) (High)79.77#266 (GPT-5.4 Nano (2026-03-17))
13Gemini 2.5 Flash Lite (Thinking)78.54#422 (Gemini 2.5 Flash Lite)
14GPT-5.277.32#110
15GPT-5 Mini (2025-08-07)76.52#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-polish-knowledge · How It Works · Data refreshed daily, snapshot 2026-10-20.