Run Meta’s Apache 2.0 Muse Glimmer 30B with llama.cpp on Windows, Mac or Linux and chat with it, including about images, in your browser with every prompt staying on your computer.
Verified sourceSource checked 11/17/2026
LOCAL
Freshness 67 out of 100. Release age: 103 days old, Aug 10, 2026. Measures age, not quality.67/ 100Freshness103 days oldRelease age · Aug 10, 2026
Popularity 72 out of 100. 149,325 downloads in 30 days; 2,168 lifetime likes. Checked Nov 21, 2026. Measures adoption, not quality.72/ 100Popularity149.3K downloads / 30d2.2K lifetime likes · Checked Nov 21, 2026
What do these scores mean?+
Two ways to compare models. Neither score measures answer quality, benchmark performance, or whether a model fits your machine.
Freshness · release age
A new release starts at 100. The score halves every 180 days: 100 × 2^(−age in days / 180). When the release date is unavailable, a repository date is a fallback—not proof of when the model was released. Missing dates stay unscored.
Downloads over the past 30 days carry 80% of the score; lifetime likes carry 20%. Both use a logarithmic scale, capped at 10 million downloads and 100,000 likes. If likes are unavailable, downloads carry all the weight. Missing download statistics stay unscored.
Scores as of Nov 21, 2026. Statistics checked Nov 21, 2026. Statistics are marked for refresh after 72 hours.Full scoring method →
Before you begin
Hardware
Estimated from Meta’s published file sizes and VRAM pairings plus runtime and context headroom. Muse Glimmer is a dense 30B model, so every parameter is used for every token: offloading layers to system RAM works but slows it down far more than it would a mixture-of-experts model. Not measured by YouRunAI.
Machine
Muse Glimmer 30B · Q4_K_M GGUF
Muse Glimmer 30B · Q4_K_XL GGUF
NVIDIA · 16 GB VRAM + 32 GB RAM
Runs with several layers in RAM; noticeably slow
Heavy offload; choose Q4_K_M instead
NVIDIA RTX 4090 · 24 GB VRAM
Recommended start (Meta’s 24 GB pairing); keep the starting context
Does not fit fully; offload a few layers or use Q4_K_M
NVIDIA RTX 5090 · 32 GB VRAM
Comfortable, with room for longer context
Recommended (Meta’s 32 GB pairing)
AMD Radeon · 24 GB VRAM
Fits; requires the HIP or Vulkan llama.cpp build on Windows (Linux tab here is NVIDIA only)
Partial offload; requires the HIP or Vulkan build
Intel Arc · 16 GB VRAM + 32 GB RAM
Partial offload through the Vulkan llama.cpp build; slow
Heavy offload; choose Q4_K_M instead
Apple Silicon · 24 GB unified memory
Too tight for this guide: the file alone is 16.8 GB
Cannot fit
Apple Silicon · 32 GB unified memory
Recommended Mac start; close other apps
Tight; use Q4_K_M at the starting context
Apple Silicon · 64 GB unified memory
Comfortable
Comfortable
CPU only · 32 GB RAM
Loads, but a dense 30B model on CPU is very slow; short tasks only
Needs 48 GB RAM or more
Both packages are Meta’s own GGUF files from meta-models/Muse-Glimmer-30B-GGUF. Meta also publishes an optional 1.6 GB DFlash drafter for speculative decoding and full-precision weights (about 64 GB of VRAM) in the base repository; this guide leaves both out to keep the setup simple.
PRIVATE CHAT · WINDOWS · MAC · LINUXSOURCE-LINKED SETUP
Chat privately with Muse Glimmer 30B
Download Meta’s GGUF once, start llama.cpp’s local server, and chat in your browser. Muse Glimmer reasons before every answer, keeps context across follow-ups, and can describe images you attach.
Choose a package Run its commands Check the result
01
BEFORE YOU BEGIN
Choose your package
Select the weights you want. The instructions, downloads and settings change with your choice.
The publisher has not provided a complete memory target for this task. Check its hardware guidance before starting.
CURRENT SETUP
Muse Glimmer 30B · Q4_K_M GGUF · official
Start at 16,384 tokens for this chat setup. The 16.8 GB file is only the weights; the runtime and context add more memory, and the vision encoder adds about 1.4 GB. Longer context needs more memory, so reduce it first if the server runs out. Estimate, not a YouRunAI measurement.
Freshness 67 out of 100. Release age: 103 days old, Aug 10, 2026. Measures age, not quality.67/ 100Freshness103 days oldRelease age · Aug 10, 2026
Popularity 72 out of 100. 149,325 downloads in 30 days; 2,168 lifetime likes. Checked Nov 21, 2026. Measures adoption, not quality.72/ 100Popularity149.3K downloads / 30d2.2K lifetime likes · Checked Nov 21, 2026
What do these scores mean?+
Two ways to compare models. Neither score measures answer quality, benchmark performance, or whether a model fits your machine.
Freshness · release age
A new release starts at 100. The score halves every 180 days: 100 × 2^(−age in days / 180). When the release date is unavailable, a repository date is a fallback—not proof of when the model was released. Missing dates stay unscored.
Downloads over the past 30 days carry 80% of the score; lifetime likes carry 20%. Both use a logarithmic scale, capped at 10 million downloads and 100,000 likes. If likes are unavailable, downloads carry all the weight. Missing download statistics stay unscored.
Set up Muse Glimmer 30B · Q4_K_M GGUF on your machine
Choose your computer first, then its GPU path. The instructions below update to match your selection.
CHOOSE YOUR SETUPWindows
01
Install llama.cpp with winget
Open PowerShell and install the official llama.cpp release with winget, then close and reopen PowerShell so llama-server is on your PATH. Meta’s card says Muse Glimmer needs llama.cpp build b10353 or newer; the second command prints your build number, and the third lists the GPUs llama.cpp can use. If only the CPU appears on an NVIDIA or AMD PC, download the CUDA or HIP Windows build from the official releases page instead and run the same commands from its folder.
In PowerShell, run this command and leave the window open. The first run downloads Meta’s 16.8 GB Muse-Glimmer-30B-KQuant-17GB-Q4_K_M.gguf file; later runs reuse it. It also downloads Meta’s 1.4 GB vision encoder so you can attach images. The server listens only on this computer at http://127.0.0.1:8080, names the model muse-glimmer-30b, and starts at 16,384 tokens of context with one request slot (-np 1), so the whole context goes to you. The sampling settings are Meta’s recommended temperature 1.0, top_p 0.95 and top_k 64, and --jinja is required for the chat template embedded in the file. Wait until the log says the server is listening.
Open http://127.0.0.1:8080 in your browser on the same computer. llama.cpp serves its own chat interface there. Start a new conversation and send the prompt below. Muse Glimmer always reasons before it answers, so you will first see a collapsible thinking section and then the reply.
Plain text
Plan a three-day study schedule for learning basic Spanish. Use one short bullet per day.
04
Ask a follow-up and attach an image
In the same conversation, send: “Make day two lighter and move the vocabulary drill to day three.” A correct reply edits the plan it already gave you. Then use the attachment button to add any photo from your computer and ask “Describe this image in two sentences.” The description should match the photo, which confirms the vision encoder loaded.
05
Use it privately, stop and restart
Once the files have downloaded, the conversation runs entirely on this computer; you can disconnect from the internet and send a new prompt to confirm. Chats are kept in this browser's storage, so use the page's delete controls on a shared computer. Press Ctrl+C in PowerShell to stop the server. To chat again, run the same command and reopen http://127.0.0.1:8080; nothing is downloaded a second time.
03
SUCCESS CHECK · Muse Glimmer 30B · Q4_K_M GGUF
A private conversation that remembers context and reads images
The chat page at http://127.0.0.1:8080 answers your first prompt, revises its own plan when you ask the follow-up, and describes a photo you attach, with the internet disconnected.
http://127.0.0.1:8080Opens on your machine
Sources checked November 17, 2026. Commands follow Meta’s published GGUF card and recommended settings; YouRunAI has not run this exact setup. Hardware ratings are estimates.
Lower -c by half, close other GPU apps, or switch to Q4_K_M. On a 16 GB GPU replace -ngl 99 with -ngl 40 to keep part of the model in system RAM; it will be slower.
llama-server says it does not recognise the model architecture.
Meta’s card says builds b10344 and older cannot load Muse Glimmer. Update llama.cpp (winget upgrade llama.cpp, brew upgrade llama.cpp, or docker pull the server-cuda image), check llama-server --version shows b10353 or newer, and try again.
Every answer starts with a thinking section, even short ones.
That is expected: Meta’s chat template always opens the reasoning channel and it cannot be switched off. To shorten it, add --reasoning-budget 1024 to the server command, or add --chat-template-kwargs '{"reasoning_strength":"low"}'.
REFERENCE LIBRARY
Sources and files
Original instructions, model files, and compatibility notes behind this setup.