BLUEX v2: leaderboard
Although Large Language Models (LLMs) excel in many tasks, their assessment in Portuguese has received less attention, particularly for open-ended, discursive tasks that demand deeper reasoning and generation capabilities. While the original BLUEX benchmark addressed the scarcity of Portuguese evaluation datasets through multiple-choice questions from Brazilian university entrance exams, it did no
Metric: Score (self-reported). Source: benchmarklist.com. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro Preview | 9.1 | |
| 2 | GPT-5.4 | 8.92 | #79 |
| 3 | GPT-5-mini | 8.9 | |
| 4 | Gemini 3.1 Flash Lite Preview | 8.83 | |
| 5 | Qwen3.5-122B-A10B | 8.83 | |
| 6 | Microsoft phi-4 | 7.09 | |
| 7 | Qwen-2.5 7b-instruct | 6.93 | |
| 8 | cohere Command-r7b-12-2024 | 5.41 | |
| 9 | Llama-3.1-8b Instruct | 4.35 | |
| 10 | Llama-3.2-11b Vision Instruct | 4.18 |
Interactive version: theaggregate.ai/benchmark?slug=bluex-v2 · How It Works · Data refreshed daily, snapshot 2026-10-20.