EuroEval Portuguese Knowledge: leaderboard

Metric: Knowledge Average Score (%). Source: euroeval.com. 396 models tracked.

Top models

#ModelScore
1GPT-593.32
2GPT-5 (High)92.7
3O3 (2025-04-16)92.18
4Grok 4.1 Fast (Reasoning)90.99
5Gemini 3 Pro (Preview)90.72
6GPT-5 Mini (High)90.72
7Gemini 2.5 Pro90.3
8Gemini 2.5 Pro (Preview 03-25)89.83
9Claude Sonnet 4.5 (Thinking)89.64
10GPT-5.4 Mini (Medium)89.58
11GPT-5 Mini89.35
12Claude Sonnet 4.688.87
13Claude Sonnet 4.5 (Non-reasoning)88.65
14Gemini 2.5 Flash (Thinking)88.56
15GPT-5.4 Mini (High)88.45

Interactive version: theaggregate.ai/benchmark?slug=euroeval-portuguese-knowledge · How It Works · Data refreshed daily, snapshot 2026-09-05.