Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

AA Global-MMLU-Lite - Portuguese — leaderboard

Metric: Accuracy (%). Source: artificialanalysis.ai. 113 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)94.25
2Claude Opus 4.6 (Adaptive Reasoning, Max Effort)93.92
3Gemini 3 Pro (Preview) (High)92.92
4Grok 492.5
5Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)92.5
6Claude Opus 4.5 (Thinking)92.42
7Grok 4.20 0309 (Reasoning)92.33
8Claude Opus 4.5 (Non-reasoning)92.17
9Gemini 2.5 Pro92.08
10GPT-5 (High)92.08
11GPT-5 (Medium)92
12Claude Sonnet 4.5 (Thinking)91.75
13GPT-5.1 (High)91.42
14Qwen 3.5 397B A17B (Non-reasoning)91.33
15GPT-5.2 (Medium)91.08

Interactive version: theaggregate.ai/benchmark?slug=aa-global-mmlu-lite-portuguese · How the rankings work · Data refreshed daily, snapshot 2026-07-22.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.