readygpu

Muse Glimmer 30B

29.8B parameters · 128k context · Apache 2.0 · released 2026-08-09

chatcodingagenticvisionreasoning

In short

Muse Glimmer 30B is an open model from meta-models with 29.8B parameters. At Q4_K_M, Muse Glimmer 30B needs about 17 GB of memory at 8k context and 17 GB at 8k, counting the weights, the KV cache and runtime buffers. The cheapest hardware in the US that runs Muse Glimmer 30B fully in memory at 8k context is the AMD Radeon RX 7900 XT, $800 at Amazon, at about 25 tok/s (estimated).

Updated . Speeds are estimates; how they are worked out.

What to run it on in the US

Fully in memory at 8k context, priced at retailers today.

All hardware that runs it

Everything we price, best value first. Two cards split the model by layers.

HardwareQuantFitsDecodePricetok/s per 1,000
NVIDIA Tesla P40Q6_Kfits9 tok/s$34526.4
AMD Radeon RX 7900 XTQ4_K_Mfits25 tok/s$80031.4
NVIDIA GeForce RTX 3090 TiQ6_Kfits27 tok/s$99526.6
Intel Arc Pro B60Q6_Kfits10 tok/s$1,0009.7
AMD Radeon RX 7900 XTXQ6_Kfits22 tok/s$1,20018.7
Mac mini M4 32GBQ5_K_Mfits4 tok/s$1,4202.7
GMKtec EVO-X2 Ryzen AI Max+ 395 128GBQ8_0fits4 tok/s$1,6002.7
NVIDIA GeForce RTX 3090Q6_Kfits25 tok/s$1,60015.4
AMD Radeon AI PRO R9700Q8_0fits11 tok/s$1,9445.6
NVIDIA GeForce RTX 4090Q6_Kfits27 tok/s$2,07312.8
Mac mini M4 Pro 48GBQ8_0fits6 tok/s$2,1502.6
Mac mini M4 Pro 64GBQ8_0fits6 tok/s$2,4902.2
Mac Studio M4 Max 64GBQ8_0fits11 tok/s$3,6003.0
NVIDIA GeForce RTX 5090Q8_0fits34 tok/s$4,9216.9
Mac Studio M3 Ultra 96GBQ8_0fits16 tok/s$4,9683.3
Mac Studio M4 Max 128GBQ8_0fits11 tok/s$5,4002.0
ASUS Ascent GX10 128GBQ8_0fits5 tok/s$6,4990.8
NVIDIA DGX Spark 128GBQ8_0fits5 tok/s$7,0500.7
Mac Studio M3 Ultra 256GBQ8_0fits16 tok/s$7,9992.1
NVIDIA RTX PRO 6000 BlackwellQ8_0fits34 tok/s$8,4904.0
Mac Studio M3 Ultra 512GBQ8_0fits16 tok/s$19,9990.8
Framework Desktop Ryzen AI Max+ 395 128GBQ8_0fits4 tok/s––
Framework Desktop Ryzen AI Max+ 395 64GBQ8_0fits4 tok/s––
Beelink GTR9 Pro Ryzen AI Max+ 395 128GBQ8_0fits4 tok/s––
Intel Arc A770 16GBQ4_K_Moffload11 tok/s$31034.5
NVIDIA GeForce RTX 3060 12GBQ4_K_Moffload6 tok/s$37016.5
Intel Arc B580Q4_K_Moffload6 tok/s$37016.2
AMD Radeon RX 9060 XT 16GBQ4_K_Moffload8 tok/s$47016.8
AMD Radeon RX 7800 XTQ4_K_Moffload12 tok/s$50023.8
NVIDIA GeForce RTX 3080 10GBQ4_K_Moffload6 tok/s$53010.9
AMD Radeon RX 7600 XTQ4_K_Moffload7 tok/s$59912.2
NVIDIA GeForce RTX 5060 Ti 16GBQ4_K_Moffload11 tok/s$61818.1
NVIDIA GeForce RTX 5070Q4_K_Moffload7 tok/s$7509.5
NVIDIA GeForce RTX 4070 SuperQ4_K_Moffload7 tok/s$7708.7
NVIDIA GeForce RTX 4060 Ti 16GBQ4_K_Moffload8 tok/s$78010.6
AMD Radeon RX 9070 XTQ4_K_Moffload12 tok/s$81014.9
NVIDIA GeForce RTX 5070 TiQ4_K_Moffload16 tok/s$99516.5
NVIDIA GeForce RTX 4070 Ti SuperQ4_K_Moffload14 tok/s$1,09912.9
NVIDIA GeForce RTX 4080 SuperQ4_K_Moffload15 tok/s$1,30011.5
NVIDIA GeForce RTX 5080Q4_K_Moffload17 tok/s$1,9008.9

1 other card or machine we price cannot hold it at this context.

Memory needed

Weights, KV cache and runtime buffers at each context length (KV cache at fp16).

QuantFile@8k@32k@64k@128k
Q4_K_Mgguf16.1 GB17.0 GB17.5 GB18.1 GB19.3 GB
mlx-4bitmlx18.1 GB19.0 GB19.5 GB20.0 GB21.2 GB
Q5_K_Mgguf18.7 GB19.6 GB20.1 GB20.7 GB21.9 GB
mlx-5bitmlx21.3 GB22.3 GB22.7 GB23.3 GB24.5 GB
Q6_Kgguf21.8 GB22.7 GB23.2 GB23.8 GB25.0 GB
mlx-6bitmlx24.6 GB25.5 GB26.0 GB26.5 GB27.7 GB
Q8_0gguf30.1 GB31.0 GB31.5 GB32.1 GB33.3 GB
mlx-8bitmlx31.1 GB32.0 GB32.5 GB33.0 GB34.2 GB

Run it

Copy a command for the runtime you use.

  • Q4_K_M
    • Ollamaollama run hf.co/bartowski/Muse-Glimmer-30B-GGUF:Q4_K_M
    • LM Studiolms get bartowski/Muse-Glimmer-30B-GGUF@q4_k_m
    • llama.cppllama-server -hf bartowski/Muse-Glimmer-30B-GGUF:Q4_K_M
  • mlx-4bit
    • LM Studiolms get mlx-community/Muse-Glimmer-30B-4bit
    • MLXmlx_lm.server --model mlx-community/Muse-Glimmer-30B-4bit
  • Q5_K_M
    • Ollamaollama run hf.co/bartowski/Muse-Glimmer-30B-GGUF:Q5_K_M
    • LM Studiolms get bartowski/Muse-Glimmer-30B-GGUF@q5_k_m
    • llama.cppllama-server -hf bartowski/Muse-Glimmer-30B-GGUF:Q5_K_M
  • mlx-5bit
    • LM Studiolms get mlx-community/Muse-Glimmer-30B-5bit
    • MLXmlx_lm.server --model mlx-community/Muse-Glimmer-30B-5bit
  • Q6_K
    • Ollamaollama run hf.co/bartowski/Muse-Glimmer-30B-GGUF:Q6_K
    • LM Studiolms get bartowski/Muse-Glimmer-30B-GGUF@q6_k
    • llama.cppllama-server -hf bartowski/Muse-Glimmer-30B-GGUF:Q6_K
  • mlx-6bit
    • LM Studiolms get mlx-community/Muse-Glimmer-30B-6bit
    • MLXmlx_lm.server --model mlx-community/Muse-Glimmer-30B-6bit
  • Q8_0
    • Ollamaollama run hf.co/bartowski/Muse-Glimmer-30B-GGUF:Q8_0
    • LM Studiolms get bartowski/Muse-Glimmer-30B-GGUF@q8_0
    • llama.cppllama-server -hf bartowski/Muse-Glimmer-30B-GGUF:Q8_0
  • mlx-8bit
    • LM Studiolms get mlx-community/Muse-Glimmer-30B-8bit
    • MLXmlx_lm.server --model mlx-community/Muse-Glimmer-30B-8bit

Notes

Dense 30B model with a ~1.8B vision encoder; 4-bit quantized variants target 24/32 GB VRAM (full precision ~64GB). Ships with a DFlash speculative decoding drafter.

Questions

How much VRAM does Muse Glimmer 30B need?
At Q4_K_M, Muse Glimmer 30B needs about 17 GB of memory at 8k context and 17 GB at 8k, counting the weights, the KV cache and runtime buffers.
What is the best GPU for running Muse Glimmer 30B locally?
The cheapest hardware in the US that runs Muse Glimmer 30B fully in memory at 8k context is the AMD Radeon RX 7900 XT, $800 at Amazon, at about 25 tok/s (estimated). The fastest we price is the NVIDIA GeForce RTX 5090, at about 34 tok/s.
Can I run Muse Glimmer 30B on a 16 GB GPU?
Only partly: at 8k context it does not fit in 16 GB, so part of it runs from system RAM and it slows to about 11 tok/s on the Intel Arc A770 16GB (estimated).
Can I run Muse Glimmer 30B on a 24 GB GPU?
Yes: at Q6_K it fits fully in 24 GB at 8k context. On the NVIDIA GeForce RTX 3090 Ti it runs at about 27 tok/s (estimated).
How do I run Muse Glimmer 30B locally?
With Ollama: ollama run hf.co/bartowski/Muse-Glimmer-30B-GGUF:Q4_K_M (the Q4_K_M build). LM Studio and llama.cpp commands are on this page.
What licence is Muse Glimmer 30B released under?
Apache 2.0, which allows commercial use.

More hardware for local AI