Meta AI Releases Muse Glimmer: A 30B Open-Weights Agentic Model That Runs on One Consumer GPU
Meta drops a 30B agentic model that runs on consumer GPUs. Open weights, Apache 2.0, 3.1x faster decoding.

Why it matters
Open-weight agentic models are democratizing agent development. A capable 30B model that fits one GPU shifts what's buildable outside the frontier labs and changes the economics of agent deployment.
The key facts
6 to knowModel size: 30B parameters
License: Apache 2.0 (fully open)
VRAM requirement: 24GB (fits consumer GPU)
Decoding speedup: 3.1x faster with DFlash speculation
Model class: agentic (agent-native architecture)
Publisher: Meta AI
Go to the source
MarkTechPostmarktechpost.com
Publisher excerpt: Meta's Muse Glimmer is a 30B open-weights agentic model under Apache 2.0. It fits 24 GB VRAM and decodes 3.1x faster with DFlash speculation.