Cohere Unveils Aya Vision: A Multilingual, Multimodal AI Model

Mar 4, 2025
Cohere For AI has launched Aya Vision, a new open-weight AI model that integrates language and vision capabilities across 23 languages. The model is available for non-commercial use on platforms like Hugging Face and WhatsApp.
Cohere Unveils Aya Vision: A Multilingual, Multimodal AI Model
Image: Cohere

Cohere For AI has introduced Aya Vision, an open-weight AI model designed to enhance multilingual and multimodal capabilities, announced on their website. Aya Vision supports 23 languages, aiming to bridge the gap in AI performance across different languages, particularly in tasks involving both text and images.

Aya Vision is available in two versions, 8-billion and 32-billion parameters, and can perform tasks such as image captioning, visual question answering, and text translation. The model is accessible for non-commercial use on platforms like Hugging Face and WhatsApp, allowing researchers and developers to explore its capabilities without commercial constraints.

Cohere emphasizes the model's efficiency, noting that Aya Vision outperforms larger models in various benchmarks. This is achieved through innovations like synthetic annotations and multimodal model merging, which enhance the model's understanding and performance across multiple languages. Aya Vision's release marks a significant step in making advanced AI technologies more accessible to the global research community.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Daily AI Brief.

Also, consider following us on social media:

AI Brief AI Brief (X)
Free newsletter

Daily AI Brief

Daily report covering major AI developments and industry news, with both top stories and complete market updates

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.