Aya Vision

Aya Vision, from Cohere For AI, is the open-weights, multilingual, multimodal models (8B & 32B). Outperforms larger models on multilingual vision tasks. Available on Hugging Face and Kaggle.

Aya Vision by Cohere For AI is a cutting-edge open-weights model designed for multilingual and multimodal AI tasks. Available in 8B and 32B versions, it excels in multilingual vision tasks, outperforming larger models. Hosted on platforms like Hugging Face and Kaggle, Aya Vision provides a robust foundation for global AI communication, enabling seamless integration of vision and language capabilities. Its open-weights approach fosters accessibility and innovation, making it a powerful tool for developers and researchers aiming to expand AI's reach across diverse linguistic and visual contexts.