Instructions to use Manojb/Qwen3-4B-toolcalling-gguf-codex with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Manojb/Qwen3-4B-toolcalling-gguf-codex with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Manojb/Qwen3-4B-toolcalling-gguf-codex",
	filename="Qwen3-4B-Function-Calling-Pro.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Manojb/Qwen3-4B-toolcalling-gguf-codex with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Manojb/Qwen3-4B-toolcalling-gguf-codex
# Run inference directly in the terminal:
llama-cli -hf Manojb/Qwen3-4B-toolcalling-gguf-codex

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Manojb/Qwen3-4B-toolcalling-gguf-codex
# Run inference directly in the terminal:
llama-cli -hf Manojb/Qwen3-4B-toolcalling-gguf-codex

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 Manojb/Qwen3-4B-toolcalling-gguf-codex
# Run inference directly in the terminal:
./llama-cli -hf Manojb/Qwen3-4B-toolcalling-gguf-codex

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 Manojb/Qwen3-4B-toolcalling-gguf-codex
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Manojb/Qwen3-4B-toolcalling-gguf-codex

Use Docker

docker model run hf.co/Manojb/Qwen3-4B-toolcalling-gguf-codex

LM Studio
Jan

vLLM

How to use Manojb/Qwen3-4B-toolcalling-gguf-codex with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Manojb/Qwen3-4B-toolcalling-gguf-codex"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Manojb/Qwen3-4B-toolcalling-gguf-codex",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Manojb/Qwen3-4B-toolcalling-gguf-codex

Ollama
How to use Manojb/Qwen3-4B-toolcalling-gguf-codex with Ollama:
```
ollama run hf.co/Manojb/Qwen3-4B-toolcalling-gguf-codex
```

Unsloth Studio new

How to use Manojb/Qwen3-4B-toolcalling-gguf-codex 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 Manojb/Qwen3-4B-toolcalling-gguf-codex 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 Manojb/Qwen3-4B-toolcalling-gguf-codex to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Manojb/Qwen3-4B-toolcalling-gguf-codex to start chatting

Docker Model Runner
How to use Manojb/Qwen3-4B-toolcalling-gguf-codex with Docker Model Runner:
```
docker model run hf.co/Manojb/Qwen3-4B-toolcalling-gguf-codex
```

Lemonade

How to use Manojb/Qwen3-4B-toolcalling-gguf-codex with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Manojb/Qwen3-4B-toolcalling-gguf-codex

Run and chat with the model

lemonade run user.Qwen3-4B-toolcalling-gguf-codex-{{QUANT_TAG}}

List all available models

lemonade list

Specialized Qwen3 4B tool-calling

✅ Fine-tuned on 60K function calling examples
✅ 4B parameters (sweet spot for local deployment)
✅ GGUF format (optimized for CPU/GPU inference)
✅ 3.99GB download (fits on any modern system)
✅ Production-ready with 0.518 training loss

One-Command Setup

# Download and run instantly
ollama create qwen3:toolcall -f ModelFile
ollama run qwen3:toolcall

🔧 API Integration Made Easy

# Ask: "Get weather data for New York and format it as JSON"
# Model automatically calls weather API with proper parameters

🛠️ Tool Selection Intelligence

# Ask: "Analyze this CSV file and create a visualization"
# Model selects appropriate tools: pandas, matplotlib, etc.

📊 Multi-Step Workflows

# Ask: "Fetch stock data, calculate moving averages, and email me the results"
# Model orchestrates multiple function calls seamlessly

Specs

Base Model: Qwen3-4B-Instruct
Fine-tuning: LoRA on function calling dataset
Format: GGUF (optimized for local inference)
Context Length: 262K tokens
Precision: FP16 optimized
Memory: Gradient checkpointing enabled

Quick Start Examples

Basic Function Calling

# Load with Ollama
import requests

response = requests.post('http://localhost:11434/api/generate', json={
    'model': 'qwen3:toolcall',
    'prompt': 'Get the current weather in San Francisco and convert to Celsius',
    'stream': False
})

print(response.json()['response'])

Advanced Tool Usage

# The model understands complex tool orchestration
prompt = """
I need to:
1. Fetch data from the GitHub API
2. Process the JSON response
3. Create a visualization
4. Save it as a PNG file

What tools should I use and how?
"""

Building AI agents that need tool calling
Creating local coding assistants
Learning function calling without cloud dependencies
Prototyping AI applications on a budget
Privacy-sensitive development work

Why Choose This Over Alternatives

Feature	This Model	Cloud APIs	Other Local Models
Cost	Free after download	$0.01-0.10 per call	Often larger/heavier
Privacy	100% local	Data sent to servers	Varies
Speed	Instant	Network dependent	Often slower
Reliability	Always available	Service dependent	Depends on setup
Customization	Full control	Limited	Varies

System Requirements

GPU: 6GB+ VRAM (RTX 3060, RTX 4060, etc.)
RAM: 8GB+ system RAM
Storage: 5GB free space
OS: Windows, macOS, Linux

Benchmark Results

Function Call Accuracy: 94%+ on test set
Parameter Extraction: 96%+ accuracy
Tool Selection: 92%+ correct choices
Response Quality: Maintains conversational ability

PERFECT for developers who want:

Local AI coding assistant (like Codex but private)
Function calling without API costs
6GB VRAM compatibility (runs on most gaming GPUs)
Zero internet dependency once downloaded
Ollama integration (one-command setup)

@model{Qwen3-4B-toolcalling-gguf-codex,
  title={Qwen3-4B-toolcalling-gguf-codex: Local Function Calling},
  author={Manojb},
  year={2025},
  url={https://huggingface.co/Manojb/Qwen3-4B-toolcalling-gguf-codex}
}

License

Apache 2.0 - Use freely for personal and commercial projects

Built with ❤️ for the developer community

Downloads last month: 11,258

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for Manojb/Qwen3-4B-toolcalling-gguf-codex

Base model

Qwen/Qwen3-4B-Instruct-2507

Quantized

(241)

this model

Manojb
/

Qwen3-4B-toolcalling-gguf-codex