Cohere releases Tiny Aya, a family of 3.35B-parameter open-weight models supporting 70+ languages for offline use, trained on a single cluster of 64 H100 GPUs
Enterprise AI company Cohere launched a new family of multilingual models on the sidelines of the ongoing India AI Summit.
Despite being smaller, Tiny Aya competes with 4B models across translation, mathematical reasoning, understanding, and generation with especially strong gains for African languages. 🌍 [image]
We take a stance for language diversity. Going beyond the one-fits-all paradigm, we release not only one instruction-finetuned model balancing all 70 languages (Tiny Aya Global), but accompany it with three region-focused models. 🌐 [image]
Introducing ✨Tiny Aya✨, a family of massively multilingual small language models built to run where people actually are. Tiny Aya delivers strong multilingual performance in 70+ global languages in a 3.35B parameter model, efficient enough to run locally, even on a phone. [video]