Mistral AI releases Pixtral 12B, a multimodal model surpassing Llama

Mistral AI has released Pixtral 12B, a new open-source multimodal model. This model is capable of processing both text and images, setting it apart from many other open-source alternatives. Pixtral 12B aims to compete with models like Google's Gemini and OpenAI's GPT-4V by offering strong performance on tasks that require understanding of visual and textual information. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

RANK_REASON Release of an open-source multimodal model from a notable AI lab.

Read on Smol AINews →

• model release
• paper