Instructions to use Moonlink/Mable-0.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Moonlink/Mable-0.5 with Transformers:

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

pipe = pipeline("text-generation", model="Moonlink/Mable-0.5")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Moonlink/Mable-0.5")
model = AutoModelForCausalLM.from_pretrained("Moonlink/Mable-0.5", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

llama-cpp-python

How to use Moonlink/Mable-0.5 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Moonlink/Mable-0.5",
	filename="gemma-2-2b-it.Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Moonlink/Mable-0.5 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 Moonlink/Mable-0.5:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Moonlink/Mable-0.5:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Moonlink/Mable-0.5:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Moonlink/Mable-0.5: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 Moonlink/Mable-0.5:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Moonlink/Mable-0.5: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 Moonlink/Mable-0.5:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Moonlink/Mable-0.5:Q4_K_M

Use Docker

docker model run hf.co/Moonlink/Mable-0.5:Q4_K_M

LM Studio
Jan

vLLM

How to use Moonlink/Mable-0.5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Moonlink/Mable-0.5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Moonlink/Mable-0.5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Moonlink/Mable-0.5:Q4_K_M

SGLang

How to use Moonlink/Mable-0.5 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 "Moonlink/Mable-0.5" \
    --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": "Moonlink/Mable-0.5",
		"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 "Moonlink/Mable-0.5" \
        --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": "Moonlink/Mable-0.5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use Moonlink/Mable-0.5 with Ollama:
```
ollama run hf.co/Moonlink/Mable-0.5:Q4_K_M
```

Unsloth Studio

How to use Moonlink/Mable-0.5 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 Moonlink/Mable-0.5 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 Moonlink/Mable-0.5 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Moonlink/Mable-0.5 to start chatting

Atomic Chat new
Docker Model Runner
How to use Moonlink/Mable-0.5 with Docker Model Runner:
```
docker model run hf.co/Moonlink/Mable-0.5:Q4_K_M
```

Lemonade

How to use Moonlink/Mable-0.5 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Moonlink/Mable-0.5:Q4_K_M

Run and chat with the model

lemonade run user.Mable-0.5-Q4_K_M

List all available models

lemonade list

Mable-0.5 🧠⚡

A high-reasoning 2B language model fine-tuned on Gemma-2-2B using Fable-5 reasoning traces.

📌 Model Overview

Mable-0.5 is a fine-tuned variant of Google's Gemma-2-2B-it, trained with 3,500 curated reasoning traces from the Fable-5 dataset. It specializes in step-by-step reasoning, structured chain-of-thought (CoT) breakdown, and execution-oriented decision making.

Developer: Moonlink
Base Model: unsloth/gemma-2-2b-it-bnb-4bit
Fine-Tuning Technique: LoRA (Rank = 16, Alpha = 32)
Optimization: Fine-tuned via Unsloth

🚀 Available Formats & Usage

This repository contains all 3 formats for maximum flexibility across deployment environments:

1. 🦙 GGUF Format (Local / Ollama / LM Studio)

Run Mable-0.5 locally on CPU or Apple Silicon using the quantized .gguf file.

Using Ollama:

# Download and run the quantized GGUF directly from Hugging Face
ollama run hf.co/Moonlink/Mable-0.5:Q4_K_M

2. ⚡ LoRA Adapters (Transformers / Unsloth)

Attach the lightweight adapter weights to the base Gemma-2-2B model.

from unsloth import FastLanguageModel

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "Moonlink/Mable-0.5",
    max_seq_length = 2048,
    load_in_4bit = True,
)
FastLanguageModel.for_inference(model)

prompt = """<start_of_turn>user
How many r's are in the word strawberry?<end_of_turn>
<start_of_turn>model
THOUGHT:
"""

inputs = tokenizer([prompt], return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

3. 📦 Merged 16-Bit Weights (vLLM / Pipeline Deployment)

Use the fully merged standalone model for production serving.

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "Moonlink/Mable-0.5"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto"
)

📋 Prompt Format

Mable-0.5 follows the Gemma chat template with explicit THOUGHT: and ACTION: structural blocks:

<start_of_turn>user
{Your prompt here}<end_of_turn>
<start_of_turn>model
THOUGHT:
{Chain-of-thought reasoning steps}

ACTION:
{Final response or action}
<end_of_turn>

🛠️ Fine-Tuning Hyperparameters

Max Sequence Length: 2,048 tokens
Optimizer: AdamW 8-bit
Learning Rate: 2e-4 (Linear decay)
Effective Batch Size: 4 (Batch size = 1, Gradient Accumulation = 4)
Epochs/Steps: 120 steps (~3,500 rows processed)
Precision: Mixed FP16/BF16

🤗 If you benefit from any of our work in HuggingFace please give us a Like or Follow.

Downloads last month: 79

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for Moonlink/Mable-0.5

Base model

unsloth/gemma-2-2b-it-bnb-4bit

Quantized

(47)

this model

Moonlink
/

Mable-0.5