Run Meta Muse Glimmer 30B on AMD Ryzen AI Max+ and Radeon GPUs
AMD, Monday, August 10th, 2026
AMD publishes a guide to running Meta's Apache 2.0 Muse Glimmer 30B model locally on AMD hardware.
AMD published a guide to running Meta Muse Glimmer 30B locally on AMD hardware. Muse Glimmer is a new 30B dense open-weights model from Meta Superintelligence Labs, released under Apache 2.0 and aimed at local agentic use cases.
AMD's argument is that agentic AI needs large context, persistent memory, tool use, privacy and low operating costs, and that cloud-first approaches introduce latency, privacy concerns and recurring token expenses.
The post shows the model running on a system powered by an AMD Ryzen AI Max+ processor or a workstation with a single AMD Radeon AI PRO R9700 graphics card, using open frameworks such as llama.cpp. AMD reports early performance testing showing strong local performance.