README.md

---
license: apache-2.0
---

## Introduction

**This repo contains the humanized 360M SmolLM2 model in the GGUF Format**
- Quantization: q2_K, q3_K_S, q3_K_M, q3_K_L, q4_0, q4_K_S, q4_K_M, q5_0, q5_K_S, q5_K_M, q6_K, q8_0

**More about this model**
- We released a 135M, 360M and 1.7B parameter version of this model. For more information, view our [report](https://www.assistantslab.com/research/smollm2-report).

## Quickstart

We advise you to clone [`llama.cpp`](https://github.com/ggerganov/llama.cpp) and install it following the official guide. We follow the latest version of llama.cpp. 
In the following demonstration, we assume that you are running commands under the repository `llama.cpp`.

Since cloning the entire repo may be inefficient, you can manually download the GGUF file that you need or use `huggingface-cli`:
1. Install
   ```shell
   pip install -U huggingface_hub
   ```
2. Download:
   ```shell
   huggingface-cli download AssistantsLab/SmolLM2-360M-humanized_GGUF smollm2-360M-humanized-q4_k_m.gguf --local-dir . --local-dir-use-symlinks False
   ```

### Quants

| Filename | Quant type | File Size |
| -------- | ---------- | --------- |
| [smollm2-1.7b-humanized-q2_k.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q2_k.gguf) | Q2_K | 675MB |
| [smollm2-1.7b-humanized-q3_k_s.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q3_k_s.gguf) | Q3_K_S | 777MB |
| [smollm2-1.7b-humanized-q3_k_m.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q3_k_m.gguf) | Q3_K_M | 860MB |
| [smollm2-1.7b-humanized-q3_k_l.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q3_k_l.gguf) | Q3_K_L | 933MB |
| [smollm2-1.7b-humanized-q4_0.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q4_0.gguf) | Q4_0 | 991MB |
| [smollm2-1.7b-humanized-q4_k_s.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q4_k_s.gguf) | Q4_K_S | 999MB |
| [smollm2-1.7b-humanized-q4_k_m.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q4_k_m.gguf) | Q4_K_M | 1.06GB |
| [smollm2-1.7b-humanized-q5_0.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q5_0.gguf) | Q5_0 | 1.19GB |
| [smollm2-1.7b-humanized-q5_k_s.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q5_k_s.gguf) | Q5_K_S | 1.19GB |
| [smollm2-1.7b-humanized-q5_k_m.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q5_k_m.gguf) | Q5_K_M | 1.23GB |
| [smollm2-1.7b-humanized-q6_k.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q6_k.gguf) | Q6_K | 1.41GB |
| [smollm2-1.7b-humanized-q8_0.gguf](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized_GGUF/blob/main/smollm2-1.7b-humanized-q8_0.gguf) | Q8_0 | 1.82GB |


## More information

For more information about this model, please visit the original model [here](https://huggingface.co/AssistantsLab/SmolLM2-1.7B-humanized).


## License

[Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)

## Citation

SmolLM2:
```bash
@misc{allal2024SmolLM2,
      title={SmolLM2 - with great data, comes great performance}, 
      author={Loubna Ben Allal and Anton Lozhkov and Elie Bakouch and Gabriel Martín Blázquez and Lewis Tunstall and Agustín Piqueres and Andres Marafioti and Cyril Zakka and Leandro von Werra and Thomas Wolf},
      year={2024},
}
```

Human-Like-DPO-Dataset:
```bash
@misc{çalık2025enhancinghumanlikeresponseslarge,
      title={Enhancing Human-Like Responses in Large Language Models}, 
      author={Ethem Yağız Çalık and Talha Rüzgar Akkuş},
      year={2025},
      eprint={2501.05032},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2501.05032}, 
}
```