LLM & Text Generation4 min reading time

AMD Releases Instella-MoE-16B-A3B: A Fully Open Mixture-of-Experts LLM With 2.8B Active Parameters Trained On Instinct GPUs

MarkTechPost
Read full post
AMD launched Instella-MoE-16B-A3B, an open Mixture-of-Experts language model with 16B parameters but only 2.8B active per token, trained on Instinct MI300X/MI325X GPUs. It includes published weights, training data, configs, and inference code under research licenses, targeting academic and enterprise research use.

More in LLM & Text Generation

DeepSeek V4.1 Flash now available on AI Gateway

Covered by 2 sources

Harvey raises $550M more to develop AI tools for legal teams

SiliconANGLE