Instructions to use ales27pm/lumen-qwen3-bootstrap-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 ales27pm/lumen-qwen3-bootstrap-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 ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

Use Docker

docker model run hf.co/ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

LM Studio
Jan

vLLM

How to use ales27pm/lumen-qwen3-bootstrap-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ales27pm/lumen-qwen3-bootstrap-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": "ales27pm/lumen-qwen3-bootstrap-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

Ollama
How to use ales27pm/lumen-qwen3-bootstrap-gguf with Ollama:
```
ollama run hf.co/ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M
```

Unsloth Studio

How to use ales27pm/lumen-qwen3-bootstrap-gguf with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for ales27pm/lumen-qwen3-bootstrap-gguf to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for ales27pm/lumen-qwen3-bootstrap-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for ales27pm/lumen-qwen3-bootstrap-gguf to start chatting

How to use ales27pm/lumen-qwen3-bootstrap-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use ales27pm/lumen-qwen3-bootstrap-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

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 "ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M" \
  --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"

Docker Model Runner
How to use ales27pm/lumen-qwen3-bootstrap-gguf with Docker Model Runner:
```
docker model run hf.co/ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M
```

Lemonade

How to use ales27pm/lumen-qwen3-bootstrap-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

Run and chat with the model

lemonade run user.lumen-qwen3-bootstrap-gguf-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use ales27pm/lumen-qwen3-bootstrap-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 ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

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 ales27pm/lumen-qwen3-bootstrap-gguf:Q4_K_M

Run Hermes

hermes

Atomic Chat

lumen-qwen3-bootstrap-gguf / README.md

ales27pm

Clarify shared-base artifact lineage scope (#2)

903587d 3 days ago

preview code

Raw

History Blame Contribute Delete

3.55 kB

	---
	language:
	- en
	library_name: llama.cpp
	pipeline_tag: text-generation
	tags:
	- gguf
	- qwen3
	- quantized
	- on-device
	- lumen
	base_model:
	- Qwen/Qwen3-1.7B
	---

	# Lumen Qwen3 1.7B shared-base GGUF

	This card documents the single shared-base artifact
	`lumen-qwen3-fast-shared-q4_k_m.gguf` in
	`ales27pm/lumen-qwen3-bootstrap-gguf`. Lumen uses it as the common base for
	role-specific LoRA adapters in its local-first iOS runtime.

	## Immutable artifact identity

	\| Field \| Value \|
	\| --- \| --- \|
	\| Lumen repository \| `ales27pm/lumen-qwen3-bootstrap-gguf` \|
	\| Lumen artifact lineage revision \| `8abae6d695408dbc75a134212dd616cd14549ae1` \|
	\| Lumen file \| `lumen-qwen3-fast-shared-q4_k_m.gguf` \|
	\| File size \| 1,282,439,264 bytes \|
	\| File SHA-256 \| `a7f6720f68f4a4567ebf7e3257041dd0b72077b518efe56890aec3516b59b9de` \|
	\| Immediate source repository \| `rippertnt/Qwen3-1.7B-Q4_K_M-GGUF` \|
	\| Immediate source revision \| `b93c0c328252f68cda622847ba0218689a8b6ba4` \|
	\| Immediate source file \| `qwen3-1.7b-q4_k_m.gguf` \|
	\| Identified upstream base \| `Qwen/Qwen3-1.7B` \|
	\| Exact upstream Qwen revision used for quantization \| Not recorded \|

	Pinned links: [Lumen artifact](https://huggingface.co/ales27pm/lumen-qwen3-bootstrap-gguf/blob/8abae6d695408dbc75a134212dd616cd14549ae1/lumen-qwen3-fast-shared-q4_k_m.gguf),
	[immediate-source artifact](https://huggingface.co/rippertnt/Qwen3-1.7B-Q4_K_M-GGUF/blob/b93c0c328252f68cda622847ba0218689a8b6ba4/qwen3-1.7b-q4_k_m.gguf),
	and [immediate-source card](https://huggingface.co/rippertnt/Qwen3-1.7B-Q4_K_M-GGUF/blob/b93c0c328252f68cda622847ba0218689a8b6ba4/README.md).

	The Lumen file is byte-identical to the immediate-source file at the pinned
	revision. Lumen only renamed and re-hosted that GGUF; it did not convert,
	quantize, merge, or otherwise modify its bytes.

	The immediate-source model card identifies `Qwen/Qwen3-1.7B` as the base,
	declares Apache-2.0, and describes GGUF conversion with llama.cpp through the
	GGUF-my-repo workflow. It does not record the immutable upstream Qwen commit
	used for the April 2025 conversion, so this card does not invent one.

	## Scope warning

	This card's byte-identity and provenance statements apply only to
	`lumen-qwen3-fast-shared-q4_k_m.gguf`. The artifact-lineage revision also
	contains separate Lumen release-bake GGUF files. Those files are not the
	byte-identical rehost documented here, and this card does not establish their
	provenance or license status. They require separate documentation before any
	repository-wide licensing claim is made.

	## Intended use

	The covered artifact is intended as a local GGUF base for a compatible
	llama.cpp runtime, including Lumen's role-adapter runtime. Model output is
	untrusted and must remain behind application-layer validation, permissions,
	approvals, privacy controls, and bounded tool execution.

	## Limitations

	- The exact upstream `Qwen/Qwen3-1.7B` revision used by the immediate-source
	quantizer is unknown.
	- Quality, safety, and performance depend on the runtime, prompt, sampling
	configuration, device, and any applied adapters.
	- This artifact is not itself an application safety or authorization system.

	## License and attribution

	The covered shared-base artifact is provided under the Apache License 2.0;
	see `LICENSE`. See `NOTICE.md` for the immutable source chain, modification
	statement, scope boundary, and attribution.

	Alibaba Cloud, the Qwen authors, rippertnt, Hugging Face, llama.cpp,
	GGUF-my-repo, and their respective contributors do not endorse Lumen, this
	rehost, or applications that use it.