Cohere launches a family of open multilingual models

Enterprise AI company Cohere launched a new family of multilingual models on the sidelines of the ongoing India AI Summit. The models, called Tiny Aya, are open-weight – meaning their underlying code is publicly available for anyone to use and modify – support more than 70 languages ​​and can run on everyday devices like laptops without the need for an internet connection.

Launched by the company’s research arm, Cohere Labs, the model supports South Asian languages ​​such as Bengali, Hindi, Punjabi, Urdu, Gujarati, Tamil, Telugu and Marathi.

The base model contains 3.35 billion parameters – a measure of its size and complexity. Cohere has also launched TinyAya-Global, a version tuned to better track user commands, for apps that require broad language support. Regional variants complete the family: TinyAya-Earth for African languages; TinyAya-Fire for South Asian Languages; and TinyAya-Water for Asia Pacific, West Asia and Europe.

“This approach allows each model to develop a stronger linguistic foundation and cultural nuance, creating systems that feel more natural and trustworthy to the communities they are intended for. At the same time, all Tiny Aya models maintain broad multilingual coverage, making them flexible starting points for further adaptation and research,” the company said in a statement.

Cohere noted that these models, which are trained on a single cluster of 64 H100 GPUs (a type of high-performance chip from Nvidia) using relatively modest computing resources, are ideal for researchers and developers building apps for native-language audiences. The models run directly on devices, so developers can use them to enable offline translations. The company noted that it built the underlying software for use on the device, requiring less computing power than most comparable models.

In linguistically diverse countries like India, this kind of offline-friendly capability can open up a diverse set of applications and use cases without the need for constant internet access.

The models are available on HuggingFace, the popular platform for sharing and testing AI models, and the Cohere Platform. Developers can download them on HuggingFace, Kaggle and Ollama for local deployment. The company is also releasing training and evaluation datasets on HuggingFace and plans to release a technical report detailing the training methodology.

The startup’s CEO Aidan Gomez said last year that the company plans to go public “soon.” According to CNBCthe company ended 2025 on a high note, with annual recurring revenue of $240 million, with 50% quarter-over-quarter growth throughout the year.