← Library · Frontier

AMD Releases Instella-MoE-16B-A3B-Think, an Open MoE Model

AMD announced the release of Instella-MoE-16B-A3B-Think, a fully open Mixture-of-Experts (MoE) language model. It features 16 billion total parameters with 2.8 billion active parameters per token. The model was trained entirely end-to-end, from pre-training through reinforcement learning.

Why it matters

This release contributes a substantial open-source MoE model from a major hardware manufacturer, potentially fostering innovation and accessibility in large language model development.

Learn one new AI thing every day.

Daily Deck sends you seven plain-English cards like this every morning. Free.

Start free