In addition to its image and text understanding, Pixtral Large has a powerful multilingual capacity. The model supports dozens of languages, including English, French, German, Spanish, Italian, Chinese, Japanese, Korean, Portuguese, Dutch, and Polish, helping to encourage innovation worldwide and promoting cultural diversity with access to models trained in multiple languages and helping to reflect different perspectives. It’s also trained on more than 80 programming languages (e.g., Python and Java), meaning it can write and understand a wide range of computer codes really well.