Instructions to use OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S
# Run inference directly in the terminal:
llama cli -hf OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S
# Run inference directly in the terminal:
llama cli -hf OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S
# Run inference directly in the terminal:
./build/bin/llama cli -hf OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Use Docker

docker model run hf.co/OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Llama new
LM Studio
Jan

vLLM

How to use OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Ollama
How to use OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF with Ollama:
```
ollama run hf.co/OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S
```
Unsloth Desktop

How to use OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF with Docker Model Runner:
```
docker model run hf.co/OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S
```

Lemonade

How to use OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Run and chat with the model

lemonade run user.Muse-Glimmer-30B-Knowledge-Stage2-GGUF-IQ1_S

List all available models

lemonade list

Hermes Agent

How to use OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF:IQ1_S" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Muse Glimmer 30B Knowledge, Stage2 GGUF (small files)

Built with Muse Spark Open. A community fine-tune, not an official Meta release.

Small GGUF files of stage 2, started from the stage 1 8x A100 adapter. The run log calls stage 2 a measurement, not the final model. This repo is private. Nothing here has had the license and personal-data review that a public release needs.

How it was made

Adapter (520,895 examples, 16,278 steps, train loss 1.10, adapter sha256 starts b568ab79) merged into BF16, then llama.cpp b11970 with an imatrix from stage 1's training text. Not made-up data.

Measured

Reworded questions about passages the model trained on, and 40 old facts. Greedy, 96 new tokens, on an A100.

File	GB	New facts	Old facts	Decode tok/s
IQ1_S	6.53	9.0% (225 of 2,500)	40 of 40	67.9
IQ1_M	7.06	11.8% (295 of 2,500)	38 of 40	58.0
IQ2_XXS	7.96	18.7% (467 of 2,500)	40 of 40	59.6

Not run on a Jetson in any log we have. The 1-bit files are close to unusable: expect loops. Larger files of this model (Q4_K_M, Q8_0, BF16) were left out for storage.

Known limits

These come from tests of the Knowledge family, and not all were repeated on this file.

It makes things up and states wrong answers as fact. Don't rely on it for facts without a source.
The training data comes from public news and web text, including publishers' pages collected through CC-NEWS. Publishing weights trained on it is an open question in our run log.
This model is not the released Knowledge model. That one is OfflineBase/Muse-Glimmer-30B-2026-Knowledge.

License

Fine-tune of Muse Glimmer 30B (meta-models/Muse-Glimmer-30B, revision cec94fcfd6512f2cfe708138c32533b346ea4f69), which Meta releases under the Apache License 2.0.

Built with Muse Spark Open. Part of the training data was written by Muse Spark Open, so this model is a Derivative under the Muse Model License. These terms are carried from OfflineBase/Muse-Glimmer-30B-2026-Knowledge-LoRA:

Keep LICENSE, MUSE_MODEL_LICENSE.md, USAGE_POLICY.md and NOTICE with every copy you share.
Show "Built with Muse Spark Open" in your product's interface, documentation, website or settings menu.
Follow the Muse Glimmer Usage Policy and the Muse Spark Open Usage Policy. Both say the models are not meant for people under 18, and both bar presenting outputs as written by a person. Tell your users the known dangers.
Keep every other condition of the Muse Model License, including section 3.

Not an official Meta release. Meta has not reviewed or endorsed it. No warranty.

Downloads last month: 107

GGUF

Model size

28B params

Architecture

muse-glimmer

Hardware compatibility

1-bit

Model tree for OfflineBase/Muse-Glimmer-30B-Knowledge-Stage2-GGUF

Base model

meta-models/Muse-Glimmer-30B

Quantized

(199)

this model