---
library_name: transformers
tags:
- unsloth
---

# llama3.2-3b-turkish-trained GGUF Quantized Models

## Technical Details
- **Quantization Tool:** llama.cpp
- **Version:** version: 5155 (64082100)

## Model Information
- **Base Model:** [yusufblbl/llama3.2-3b-turkish-trained](https://huggingface.co/yusufblbl/llama3.2-3b-turkish-trained)
- **Quantized by:** [matrixportal](https://huggingface.co/matrixportal)

## Available Files
| 🚀 Download | 🔢 Type | 📝 Description |
|------------|---------|---------------|
| [Download](https://huggingface.co/matrixportal/llama3.2-3b-turkish-trained-GGUF/resolve/main/llama3.2-3b-turkish-trained.q5_k_m.gguf) | Q5 K M | 5-bit best (recommended HQ option) |
| [Download](https://huggingface.co/matrixportal/llama3.2-3b-turkish-trained-GGUF/resolve/main/llama3.2-3b-turkish-trained.q8_0.gguf) | Q8 0 | 8-bit maximum (overkill for most) |
| [Download](https://huggingface.co/matrixportal/llama3.2-3b-turkish-trained-GGUF/resolve/main/llama3.2-3b-turkish-trained.f16.gguf) | F16 | Full precision (maximum accuracy) |

💡 **Q4 K M** provides the best balance for most use cases