README.md

metadata

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.

Quickstart

We advise you to clone 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

   pip install -U huggingface_hub
   

2. Download:

   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	Q2_K	675MB
smollm2-1.7b-humanized-q3_k_s.gguf	Q3_K_S	777MB
smollm2-1.7b-humanized-q3_k_m.gguf	Q3_K_M	860MB
smollm2-1.7b-humanized-q3_k_l.gguf	Q3_K_L	933MB
smollm2-1.7b-humanized-q4_0.gguf	Q4_0	991MB
smollm2-1.7b-humanized-q4_k_s.gguf	Q4_K_S	999MB
smollm2-1.7b-humanized-q4_k_m.gguf	Q4_K_M	1.06GB
smollm2-1.7b-humanized-q5_0.gguf	Q5_0	1.19GB
smollm2-1.7b-humanized-q5_k_s.gguf	Q5_K_S	1.19GB
smollm2-1.7b-humanized-q5_k_m.gguf	Q5_K_M	1.23GB
smollm2-1.7b-humanized-q6_k.gguf	Q6_K	1.41GB
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.

License

Apache 2.0

Citation

SmolLM2:

@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:

@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}, 
}