Pixtral Large

Mistral AI🇫🇷 France
active

Mistral's frontier vision model. Based on Mistral Large 2.

Version History

largemajor

Pixtral Large is Mistral's new multimodal model, significantly more powerful than Pixtral 12B for document and image understanding. Powers le Chat's document analysis capabilities.

1.0major

Mistral AI releases Pixtral Large, a new multimodal model supporting image and text inputs with 128K context window.

pixtral-large-2411major

Pixtral Large brings frontier vision to Mistral. Built on Mistral Large 2 backbone with native multimodal understanding of images and documents.

Benchmark Scores

Full leaderboard →
93.3%
DocVQA
64.0%
MMMU

Coverage