Mistral, a French AI startup, has launched Pixtral Large, a 124-billion-parameter multimodal AI model. The model is available for non-commercial use through Hugging Face, while commercial use requires access through Mistral's API or a separate license.
Pixtral Large features a 123-billion-parameter decoder and a 1-billion-parameter vision encoder, with a context window of 128,000 tokens. The company claims the model can process up to 30 high-resolution images per input and handle multilingual OCR, reasoning, chart understanding, and document analysis.
Analyst QuickTake: This move aligns with the growing demand for multimodal AI capabilities, where users expect integration of text, image, and video generation. This launch follows its previous multimodal model launch, Pixtral 12 billion , in September. By offering Pixtral, Mistral diversifies its offerings while tapping into a lucrative market segment that has seen growth, particularly with the success of models like DALL-E and Midjourney.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.