Instructions to use Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- llama-cpp-python
How to use Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf", filename="Qwen3-4B.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 Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf # Run inference directly in the terminal: llama-cli -hf Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf # Run inference directly in the terminal: llama-cli -hf Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
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 Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf # Run inference directly in the terminal: ./llama-cli -hf Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
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 Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf # Run inference directly in the terminal: ./build/bin/llama-cli -hf Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
Use Docker
docker model run hf.co/Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
- LM Studio
- Jan
- vLLM
How to use Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-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": "Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
- Ollama
How to use Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf with Ollama:
ollama run hf.co/Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
- Unsloth Studio
How to use Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-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 Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-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 Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf to start chatting
- Pi
How to use Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
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": "Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
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 Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
Run Hermes
hermes
- Docker Model Runner
How to use Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf with Docker Model Runner:
docker model run hf.co/Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
- Lemonade
How to use Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
Run and chat with the model
lemonade run user.Qwen3-4B-Instruct-2507-rewriter-gguf-{{QUANT_TAG}}List all available models
lemonade list
Qwen3-4B-Instruct Text Rewriter (GGUF Format)
Example Notebook
๐Kaggle notebook here
Model Description
This model is a fine-tuned version of Qwen3-4B-Instruct-2507, optimized specifically to rewrite AI-generated text into a more natural, human-like style. It was trained to reduce the "robotic" tone often found in LLM outputs while maintaining the original meaning and factual accuracy.
- Base Model: unsloth/Qwen3-4B-Instruct-2507 (merged from
unsloth/Qwen3-4B-Instruct-2507-bnb-4bit) - Fine-tuned by: Heatw4ve
- Fine-tuning Tool: Unsloth
- Quantization format: GGUF (Quantized to Q4_K_M)
- Training framework: Unsloth (LoRA-based SFT)
- Language: English
- Intended usage: Local inference, research, and experimentation
Training Details
The model was fine-tuned using LoRA (Low-Rank Adaptation) on a custom dataset of 28,000+ pairs of AI vs. Human-rewritten text.
Training Configuration
The model was fine-tuned with the following parameters:
- Epochs: 2
- Learning Rate: 5e-5
- Optimizer: AdamW 8-bit
- LR Scheduler: Cosine
- Warmup Steps: 100
- Context Length: 512 tokens
- LoRA Config: - Rank (r): 64
- Alpha: 128
- Target Modules: q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
- Effective Batch Size: 128 (32 per device * 4 grad accumulation)
Usage
This is a GGUF version, compatible with LM Studio, Ollama, and llama.cpp. Works best with inputs of length 15-150 words.
System Prompt
To achieve the intended rewriting style, use the following system prompt:
You are a helpful assistant that rewrites AI-toned text into natural, human-like writing.
User Prompt Format
Rewrite the following text to sound like a real human wrote it:
[INSERT TEXT HERE]
Example Code
from llama_cpp import Llama
llm = Llama.from_pretrained(
repo_id="Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf",
filename="Qwen3-4B.gguf",
n_ctx=4096,
n_gpu_layers=16,
n_threads=4,
verbose=False,
chat_format="chatml",
)
text='''
William Shakespeare, the master of human insight, gifted us with a timeless observation that cuts to the heart of intellect and humility: "The fool doth think he is wise, but the wise man knows himself to be a fool." Uttered by the character Touchstone in As You Like It, this seemingly paradoxical statement is far more than a witty quip; it's a profound commentary on self-awareness, the nature of true wisdom, and the perpetual quest for knowledge.
'''
messages = [
{"role": "system", "content": "You are a helpful assistant that rewrites AI-toned text into natural, human-like writing."},
{"role": "user", "content": f"Rewrite the following text to sound like a real human wrote it:\n\n{text}"}
]
output = llm.create_chat_completion(
messages=messages,
max_tokens=512,
temperature=random.uniform(0.9, 1.2),
top_p=random.uniform(0.87, 0.96),
seed=random.randint(2, 2**32),
stop=["<|im_end|>", "<|endoftext|>"]
)
print(output["choices"][0]["message"]["content"].strip())
๐Connect with me on LinkedIn
- Downloads last month
- 18
We're not able to determine the quantization variants.
Model tree for Heatw4ve/Qwen3-4B-Instruct-2507-rewriter-gguf
Base model
Qwen/Qwen3-4B-Instruct-2507