Feb 17, 2026
Tiny Aya: Bridging Scale and Multilingual Depth
Tiny Aya, a compact 3.35B-parameter multilingual model, achieves state-of-the-art translation and understanding across 70 languages through innovative training and region-specialized variants, offering an efficient, balanced path for practical multilingual AI deployment.
Authors
Alejandro R. Salamanca, Diana Abagyan, Daniel D’souza, Ammar Khairi, David Mora, Saurabh Dash, Viraat Aryabumi, Sara Rajaee, Mehrnaz Mofakhami, Ananya Sahu, Thomas Euyang, Brittawnya Prince, Madeline Smith, Hangyu Lin, Acyr Locatelli, Sara Hooker, Tom Kocmi, Aidan Gomez, Ivan Zhang, Phil Blunsom, Nick Frosst, Joelle Pineau, Beyza Ermis, Ahmet Üstün, Julia Kreutzer and Marzieh Fadaee
Abstract
Tiny Aya redefines what a small multilingual language model can achieve. Trained on 70 languages and refined through region-aware posttraining, it delivers state-of-the-art in translation quality, strong multilingual understanding, and high-quality target-language generation, all with just 3.35B parameters. The release includes a pretrained foundation model, a globally balanced instruction-tuned variant, and three region-specialized models targeting languages from Africa, South Asia, Europe, Asia-Pacific, and West Asia. This report details the training strategy, data composition, and comprehensive evaluation framework behind Tiny Aya, and presents an alternative scaling path for multilingual AI: one centered on efficiency, balanced performance across languages, and practical deployment.
Related works
Research
Findings of the WMT25 Multilingual Instruction Shared Task: Persistent Hurdles in Reasoning, Generation, and Evaluation
Read
Research
Unlocking Reasoning Capability on Machine Translation in Large Language Models
Read
Research
EAGER: Entropy-Aware Generation for Adaptive Inference-Time Scaling
Read