README.md

---
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/zooai/coder-1-gguf/blob/main/LICENSE
pipeline_tag: text-generation
tags:
  - zoo
  - coder
  - coding
  - a3b
  - gguf
  - quantized
---

# Zoo Coder-1 GGUF (Quantized Coding Model)

<a href="https://zoo.ngo/" target="_blank" style="margin: 2px;">
    <img alt="Zoo AI" src="https://img.shields.io/badge/💻%20Zoo%20Coder--1%20-EF4444" style="display: inline-block; vertical-align: middle;"/>
</a>
<a href="https://zoo.ngo/" target="_blank" style="margin: 2px;">
    <img alt="501(c)(3)" src="https://img.shields.io/badge/501(c)(3)-Nonprofit-blue" style="display: inline-block; vertical-align: middle;"/>
</a>

## Overview

**Zoo Coder-1 GGUF** provides quantized versions of our enterprise-grade coding AI model. These GGUF-formatted models enable efficient deployment across various hardware configurations while maintaining excellent coding capabilities.

## Model Details

- **Base**: Qwen3-Coder with A3B technology  
- **Format**: GGUF quantized
- **Context**: 32K tokens (extensible to 128K)
- **Languages**: Python, JavaScript, TypeScript, Go, Rust, Java, C++, and 50+ more

## Available Quantizations

| Variant | Size | RAM Required | Use Case |
|---------|------|--------------|----------|
| Q2_K | ~2GB | 4GB | Edge devices, prototyping |
| Q3_K_M | ~2.5GB | 5GB | Mobile, lightweight servers |
| Q4_K_M | ~3.2GB | 6GB | **Recommended** - Best balance |
| Q5_K_M | ~4GB | 7GB | High-quality production |
| Q6_K | ~5GB | 8GB | Maximum quality |

## Quick Start

### With llama.cpp
```bash
./main -m Q4_K_M-GGUF/Q4_K_M-GGUF-00001-of-00032.gguf \
  -p "Write a Python function to calculate fibonacci numbers"
```

### With Zoo Desktop
```bash
zoo model download coder-1-gguf
```

## About Zoo AI

Zoo Labs Foundation Inc is a 501(c)(3) nonprofit organization pioneering accessible AI infrastructure.

- **Website**: [zoo.ngo](https://zoo.ngo)
- **HuggingFace**: [huggingface.co/zooai](https://huggingface.co/zooai)

## License

Apache 2.0