AMD Launches Fully Open-Source Instella-MoE-16B Model

AMD has entered the open-source large language model field by launching the Instella-MoE-16B-A3B, a Mixture of Experts model with 16 billion total and 2.8 billion active parameters, trained entirely on AMD Instinct GPUs.

2026-08-03 ~ 2026-08-03 · 2 related posts

• AMD Enters Open-Source LLM Arena with Instella-MoE-16B — airesearch12 · 2026-08-03
• AMD Releases Instella-MoE-16B: A Fully Open Mixture-of-Experts LLM — mpuchala · 2026-08-03