Yes — with 33.7 GB to spare
Muse Glimmer 30B at Q4_K_M fits your CPU only · DDR5 dual-channel entirely on the GPU at 8K context, at an estimated 2.3 tokens per second. There is room for its full 128K window.
Fully on GPU
8K context
Q4_K_M · 16.8 GB
Apache 2.0
Released 10 Aug 2026
New this month
Vision
Meta's first open weights since Llama 4: a dense 30B distilled from Muse Spark for always-on local agents. Two KV heads keep the cache small.
The VRAM budget
weights 16.8 GB
Weights 16.8 GB
KV cache @ 8K 0.18 GB
Runtime overhead 0.6 GB
Free 33.7 GB of 51.2 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 29.5 GB | 30.3 GB | 128K | 1.3 | −0.1% ppl | Long context |
| Q6_K | 22.8 GB | 23.5 GB | 128K | 1.7 | −0.4% ppl | Long context |
| Q5_K_M | 19.7 GB | 20.4 GB | 128K | 1.9 | −0.8% ppl | Long context |
| Q4_K_M | 16.8 GB | 17.5 GB | 128K | 2.3 | −1.9% ppl | Recommended |
| Q3_K_M | 13.6 GB | 14.3 GB | 128K | 2.8 | −5.4% ppl | Long context |
| Q2_K | 11.6 GB | 12.4 GB | 128K | 3.2 | −15% ppl | Long context |
Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it. This model interleaves sliding-window layers (2048 tokens, 13 of 52 layers global), which is why its cache barely grows with context.
How to run it
$ llama-server \
-hf meta-models/Muse-Glimmer-30B:Q4_K_M \
-c 8192 -ngl 99
The engine underneath most of the others. Every knob is exposed. More on llama.cpp.
01Download is 16.8 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02With no GPU, thread count matters more than clock. Start at one thread per physical core.
03There is room to go to the model's full 128K context on this card.