Instructions to use moonlight-labs/moonlight-qwen3-4b-v7 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use moonlight-labs/moonlight-qwen3-4b-v7 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="moonlight-labs/moonlight-qwen3-4b-v7")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("moonlight-labs/moonlight-qwen3-4b-v7")
model = AutoModelForCausalLM.from_pretrained("moonlight-labs/moonlight-qwen3-4b-v7", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use moonlight-labs/moonlight-qwen3-4b-v7 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 moonlight-labs/moonlight-qwen3-4b-v7:Q8_0
# Run inference directly in the terminal:
llama cli -hf moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf moonlight-labs/moonlight-qwen3-4b-v7:Q8_0
# Run inference directly in the terminal:
llama cli -hf moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

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 moonlight-labs/moonlight-qwen3-4b-v7:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

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 moonlight-labs/moonlight-qwen3-4b-v7:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

Use Docker

docker model run hf.co/moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

Llama new
LM Studio
Jan

vLLM

How to use moonlight-labs/moonlight-qwen3-4b-v7 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "moonlight-labs/moonlight-qwen3-4b-v7"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "moonlight-labs/moonlight-qwen3-4b-v7",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

SGLang

How to use moonlight-labs/moonlight-qwen3-4b-v7 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "moonlight-labs/moonlight-qwen3-4b-v7" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "moonlight-labs/moonlight-qwen3-4b-v7",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "moonlight-labs/moonlight-qwen3-4b-v7" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "moonlight-labs/moonlight-qwen3-4b-v7",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use moonlight-labs/moonlight-qwen3-4b-v7 with Ollama:
```
ollama run hf.co/moonlight-labs/moonlight-qwen3-4b-v7:Q8_0
```
Unsloth Desktop

How to use moonlight-labs/moonlight-qwen3-4b-v7 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

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": "moonlight-labs/moonlight-qwen3-4b-v7:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use moonlight-labs/moonlight-qwen3-4b-v7 with Docker Model Runner:
```
docker model run hf.co/moonlight-labs/moonlight-qwen3-4b-v7:Q8_0
```

Lemonade

How to use moonlight-labs/moonlight-qwen3-4b-v7 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

Run and chat with the model

lemonade run user.moonlight-qwen3-4b-v7-Q8_0

List all available models

lemonade list

Hermes Agent

How to use moonlight-labs/moonlight-qwen3-4b-v7 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

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 moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use moonlight-labs/moonlight-qwen3-4b-v7 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf moonlight-labs/moonlight-qwen3-4b-v7:Q8_0

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 "moonlight-labs/moonlight-qwen3-4b-v7:Q8_0" \
  --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"

🌙 Moonlight Qwen3-4B v7

Moonlight is a privacy-first, sovereign AI assistant fine-tuned on top of Qwen3-4B. It is designed specifically for local and edge execution on consumer laptops (RTX 2050 4GB VRAM) and Android smartphones (via GGUF).

🌟 Key Features

Strict Anti-Fabrication & Honesty: Refuses to invent facts, benchmark numbers, false status reports, or imaginary tool capabilities.
Privacy-First Sovereignty: Engineered for 100% offline, local execution. Zero telemetry, zero cloud data transfer.
Fluent Roman Hinglish: Native support for conversational Roman Hindi mixed with English without verbose repetitive boilerplate.
Tool Honesty & Edge Grounding: Clear operational boundaries distinguishing internal knowledge from verified external runtime inputs.
4GB GPU & Mobile Friendly: Runs locally in 4-bit NF4 with CPU offload on 4GB GPUs, or in Q4_K_M / Q8_0 GGUF on Android devices.

📊 Benchmark & Evaluation Results

Tested against the frozen Moonlight Generalization Holdout Benchmark (36 test items):

Evaluation Dimension	Pass Rate	Details
Python Code Generation	100% (4 / 4)	Passed all executable unit tests and edge cases
Tool & Boundary Honesty	100% (4 / 4)	Strict refusal of ungranted filesystem/network access
Adversarial Pressure	Pass	Refuses user pressure to mark unconfirmed operations as successful
Provenance & Identity	100% (4 / 4)	Accurate attribution of engine vs publisher
Zero Regressions	Verified	No baseline capabilities broken from previous champion versions

🚀 Quickstart Usage

Using Hugging Face Transformers

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig

model_id = "your-username/moonlight-qwen3-4b-v7"

# 4-bit Quantization (fits in 3.5 GB VRAM)
quant_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.float16,
    bnb_4bit_use_double_quant=True,
)

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    quantization_config=quant_config,
    device_map="auto",
    torch_dtype=torch.float16,
)

system_prompt = """You are Moonlight, a privacy-first AI assistant powered by Qwen3-4B.
Be accurate, helpful, practical, and concise.
Never invent facts, sources, developer information, ownership information, benchmark results, measurements, or tool access."""

messages = [
    {"role": "system", "content": system_prompt},
    {"role": "user", "content": "What are the core design principles of Moonlight AI?"},
]

inputs = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to(model.device)

outputs = model.generate(inputs, max_new_tokens=256, do_sample=False)
response = tokenizer.decode(outputs[0][inputs.shape[1]:], skip_special_tokens=True)
print(response)

📱 Mobile & Edge Deployment (GGUF)

Moonlight can be quantized and executed offline on Android using llama.cpp, Ollama, or PocketPal AI:

# Run with Ollama
ollama create moonlight -f Modelfile
ollama run moonlight

📜 Model Details

Model Name: Moonlight Qwen3-4B v7
Base Architecture: Qwen3-4B (Dense Causal LM)
Parameters: 4.02 Billion
Context Length: Up to 40,960 tokens
Training Method: 4-bit QLoRA SFT with layer streaming
License: Apache 2.0

Downloads last month: 53

Safetensors

Model size

4B params

Tensor type

F16

Model tree for moonlight-labs/moonlight-qwen3-4b-v7

Base model

Qwen/Qwen3-4B-Base

Finetuned

Qwen/Qwen3-4B

Quantized

(360)

this model

Quantizations

2 models