P3B3: leaderboard
As Large Language Models (LLMs) become embedded in everyday communication, capturing regional linguistic variation is essential for reliable and equitable language use. In Portuguese, European (pt-PT) and Brazilian (pt-BR) varieties remain unevenly represented, with pt-BR dominating in data quantity, while LLM preference for Portuguese variants remains underexplored. To address this gap, we introd
Metric: LLM pt-PT (↑) (self-reported). Source: benchmarklist.com. Saturation forecast: Around July 2027. 20 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 99.8 | #99 |
| 2 | Gemma 4 31B | 98 | #200 |
| 3 | Gemma 4 E4B | 94.4 | #767 |
| 4 | Qwen 3.5 27B | 93.3 | #187 |
| 5 | Gemma 3 12B | 92.9 | #671 |
| 6 | Ministral 3 14B | 90.7 | #650 |
| 7 | Qwen 3.5 9B | 86.1 | #379 |
| 8 | Llama 3.3 70B | 67.8 | #579 |
| 9 | salamandra-7B | 52.8 | #1461 |
| 10 | Llama 3.1 8B | 47.7 | #1151 |
| 11 | Qwen 3 8B | 32.7 | #674 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=p3b3 · How It Works · Data refreshed daily, snapshot 2026-10-20.