Muse Glimmer 30B
Muse Glimmer 30B loads in 16.9 GB at Q4_K_M and is comfortable from 32 GB of memory, Macs included from the MacBook Pro M5 Pro 48GB. Below: the full memory math, the cheapest card that runs it well, and the fastest machine we track.
PARAMETERS
29.8B
FORMAT
Q4_K_M
MIN MEMORY
24 GB
BEST FOR
Chat
Memory math
Weights (Q4_K_M)
16.9 GB
+ KV cache (16k)
~4.0 GB
Total at 16k
~20.9 GB
Comfortable from
32 GB
KV = fp16 estimate (q8_0 cache roughly halves it). "Comfortable" = weights + KV within the engine's tiered budget (~70-85% of memory).
Hardware snapshot
Frequently asked questions
How much memory does Muse Glimmer 30B need?
16.9 GB for the Q4_K_M weights, plus ~4.0 GB of KV-cache at 16k context — about 20.9 GB total. Comfortable from 32 GB of VRAM or unified memory.
Does Muse Glimmer 30B run on a Mac?
Yes — from the MacBook Pro M5 Pro 48GB (~13 tok/s est.). Unified memory means the RAM budget is the only limit.
What is the cheapest GPU for Muse Glimmer 30B?
The AMD Radeon RX 7900 XT is the cheapest tracked card that runs Muse Glimmer 30B comfortably — ~24 tok/s est. at ~$550 used.
Cite this page
ModelFit: Muse Glimmer 30B — specs, memory math and hardware verdicts. https://modelfit.io/models/muse-glimmer-30b/ (dataset updated 2026-09-03, CC BY 4.0).