---
base_model: user/merged
base_model_relation: quantized
library_name: gguf
pipeline_tag: text-generation
tags:
- gguf
- quantized
- "base_model:quantized:user/merged"
- iq4_nl
- iq4_nl_gguf
- iq4_xs
- iq4_xs_gguf
- q4_k_m
- q4_k_m_gguf
license: apache-2.0
---

# GGUF quantizations (multi-variant repo)

This repository bundles several **GGUF** exports of the same fine-tuned model, differing only by quantization method.
Lineage: **quantized** from [`user/merged`](https://huggingface.co/user/merged).

Use with llama.cpp, Ollama, LM Studio, or other GGUF-capable runtimes. Pick a subfolder that matches your hardware.

## Variants

| Folder | Quantization |
|--------|--------------|
| `iq4_nl/` | iq4_nl |
| `iq4_nl_gguf/` | iq4_nl_gguf |
| `iq4_xs/` | iq4_xs |
| `iq4_xs_gguf/` | iq4_xs_gguf |
| `q4_k_m/` | q4_k_m |
| `q4_k_m_gguf/` | q4_k_m_gguf |

## Paths on the Hub

- [`iq4_nl/`](https://huggingface.co/user/bundle/tree/main/iq4_nl)
- [`iq4_nl_gguf/`](https://huggingface.co/user/bundle/tree/main/iq4_nl_gguf)
- [`iq4_xs/`](https://huggingface.co/user/bundle/tree/main/iq4_xs)
- [`iq4_xs_gguf/`](https://huggingface.co/user/bundle/tree/main/iq4_xs_gguf)
- [`q4_k_m/`](https://huggingface.co/user/bundle/tree/main/q4_k_m)
- [`q4_k_m_gguf/`](https://huggingface.co/user/bundle/tree/main/q4_k_m_gguf)