Showing 3 results for "Multimodal"

Mistral AI has launched its first multimodal model, Pixtral 12B, offering powerful vision-and-text capabilities under an open license.

Meta's latest release features multimodal capabilities and small models optimized for mobile and wearable devices.

Google's Gemini 1.5 Pro features a massive 2-million-token context window, allowing the model to analyze entire libraries and hours of video in one go.
End of Collection