All terms
Models & Products
Pixtral
Mistral AI's multimodal model family that handles both images and text.
Definition
Pixtral is Mistral AI's family of multimodal models, which process both images and text within a single model. They extend Mistral's language models with a vision component, allowing the model to answer questions about images, read charts and documents, and combine visual and textual reasoning. Pixtral fits into Mistral's broader lineup as its vision-language offering.