Outmake Models
Models

Muse Glimmer 30B

meta-models · August 2026

Get Hugging Face

A 28B chat model from meta-models that reads pictures as well as text.

Rating
3.8
Downloads
1.3M
Likes
1.1K
Parameters
28B
Context
128K
Licence
apache-2.0

Runs well on M4 · 16 GB

The IQ2_S size from bartowski is the one to get: a 10 GB download that uses 10 GB of memory, and writes about 7 words a second.

Ratings

3.8 out of 5, across the tests below.

ScienceGPQA Diamond
83.5%
Expert questionsHumanity's Last Exam
22%
Software fixesSWE-bench Verified
76%
MathsAIME 2026
94.7%

GPQA Diamond: questions PhD scientists write for each other. Humanity's Last Exam: questions written to be beyond today's models. SWE-bench Verified: real GitHub issues fixed so the project's tests pass. AIME 2026: this year's AIME competition problems. Scores are the ones filed on the model's Hugging Face page or the board's own table, read 2026-11-05; some come from the maker.

Muse Glimmer 30B runs well from a Mac with an M1 and 16 GB of memory. 56 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From lmstudio-community/Muse-Glimmer-30B-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q4_K_M17 GB17 GBestimatedOpen

Run It Anywhere

Muse Glimmer 30B is an open model, so any app that runs GGUF files can run it. These take the IQ2_S size from bartowski straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/bartowski/Muse-Glimmer-30B-GGUF:IQ2_S
  • llama.cppMac, Windows, Linuxllama-server -hf bartowski/Muse-Glimmer-30B-GGUF:IQ2_S
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/bartowski/Muse-Glimmer-30B-GGUF:IQ2_S
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 17 GB
8 GBNone worth runningneeds 17 GB
10 GBNone worth runningneeds 17 GB
11 GBNone worth runningneeds 17 GB
12 GBNone worth runningneeds 17 GB
16 GBNone worth runningneeds 17 GB
20 GBQ4_K_M17 GB17 GB
24 GBQ4_K_M17 GB17 GB
32 GBQ4_K_M17 GB17 GB
48 GBQ4_K_M17 GB17 GB

The largest size, of those that keep three bits a weight or more, whose memory at an 8K chat is under 92% of the card's; the rest is the desktop's. Estimated from the files, the same way as for a Mac; speed on a PC is not estimated yet. A size that doesn't fit the card still runs, slower, with part of it in ordinary memory.

On every Mac
MacSizeFitMemoryWrites
M6 · 16 GBIQ2_SRuns well10 GB9.5
M6 · 24 GBIQ4_XSRuns well16 GB6.2
M6 · 32 GBUD-IQ3_MRuns great14 GB6.8
M5 · 16 GBIQ2_SRuns well10 GB8.6
M5 · 24 GBIQ4_XSRuns well16 GB5.6
M5 · 32 GBUD-IQ3_MRuns great14 GB6.1
M5 Pro · 24 GBIQ4_XSRuns well16 GB11
M5 Pro · 48 GBUD-Q5_K_XLRuns great22 GB8
M5 Pro · 64 GBQ8_0Runs great30 GB7.1
M5 Max · 36 GBQ4_K_MRuns great17 GB15
M5 Max · 48 GBUD-Q5_K_XLRuns great22 GB16
M5 Max · 64 GBQ8_0Runs great30 GB14
M5 Max · 128 GBQ8_0Runs great30 GB14
M5 Ultra · 96 GBQ8_0Runs great30 GB28
M5 Ultra · 256 GBQ8_0Runs great30 GB28
M5 Ultra · 512 GBQ8_0Runs great30 GB28
M4 · 16 GBIQ2_SRuns well10 GB7
M4 · 24 GBIQ4_XSRuns well16 GB4.6
M4 · 32 GBUD-IQ3_MRuns great14 GB5
M4 Pro · 24 GBIQ4_XSRuns well16 GB10
M4 Pro · 48 GBUD-Q5_K_XLRuns great22 GB7.4
M4 Pro · 64 GBQ8_0Runs great30 GB6.2
M4 Max · 36 GBQ4_K_MRuns great17 GB14
M4 Max · 48 GBUD-Q5_K_XLRuns great22 GB15
M4 Max · 64 GBQ8_0Runs great30 GB12
M4 Max · 96 GBQ8_0Runs great30 GB9.3
M4 Max · 128 GBQ8_0Runs great30 GB12
M3 · 8 GBIQ2_XXSToo big9.0 GB
M3 · 16 GBIQ2_SRuns well10 GB5.9
M3 · 24 GBIQ4_XSRuns well16 GB3.8
M3 Pro · 18 GBUD-IQ2_XSRuns well12 GB7.7
M3 Pro · 36 GBQ4_K_MRuns great17 GB5.3
M3 Max · 36 GBQ4_K_MRuns great17 GB10
M3 Max · 48 GBUD-Q5_K_XLRuns great22 GB11
M3 Max · 64 GBQ8_0Runs great30 GB9.1
M3 Max · 96 GBQ8_0Runs great30 GB6.8
M3 Max · 128 GBQ8_0Runs great30 GB9.1
M3 Ultra · 96 GBQ8_0Runs great30 GB18
M3 Ultra · 256 GBQ8_0Runs great30 GB18
M3 Ultra · 512 GBQ8_0Runs great30 GB18
M2 · 8 GBIQ2_XXSToo big9.0 GB
M2 · 16 GBIQ2_SRuns well10 GB5.9
M2 · 24 GBIQ4_XSRuns well16 GB3.8
M2 Pro · 16 GBIQ2_SRuns well10 GB12
M2 Pro · 32 GBUD-IQ3_MRuns great14 GB8.3
M2 Max · 32 GBUD-IQ3_MRuns great14 GB17
M2 Max · 64 GBQ8_0Runs great30 GB9.1
M2 Max · 96 GBQ8_0Runs great30 GB9.1
M2 Ultra · 64 GBQ8_0Runs great30 GB18
M2 Ultra · 128 GBQ8_0Runs great30 GB18
M2 Ultra · 192 GBQ8_0Runs great30 GB18
M1 · 8 GBIQ2_XXSToo big9.0 GB
M1 · 16 GBIQ2_SRuns well10 GB4
M1 Pro · 16 GBIQ2_SRuns well10 GB12
M1 Pro · 32 GBUD-IQ3_MRuns great14 GB8.3
M1 Max · 32 GBUD-IQ3_MRuns great14 GB16
M1 Max · 64 GBQ8_0Runs great30 GB9
M1 Ultra · 64 GBQ8_0Runs great30 GB18
M1 Ultra · 128 GBQ8_0Runs great30 GB18

Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.

Packaged by

8 repositories on Hugging Face carry this model for llama.cpp. Outmake reads them all as one model.

Information

Maker
meta-models
Released
August 2026
Parameters
28B
Longest chat
128K tokens
Reads pictures
Yes
Licence
apache-2.0
Downloads this month
1,270,659
Likes
1,058
Sizes
1, 17 to 17 GB
Kind
Maker's release

Badge for the model card

Shows on Muse Glimmer 30B's Hugging Face card which Macs run it well, linked to this page. It updates every morning.

Runs on Mac badge

Files, downloads and likes read from Hugging Face on November 5, 2026 and refreshed every morning. Memory and speed are estimates until someone running Outmake measures this model on that Mac.

Your Computer

Every model is judged for this Mac. Both are under About This Mac, in the Apple menu.

Every Mac · Every graphics card

Civitai

Open in OutmakeHugging Face

New to Outmake? Download the free beta