---
base_model: unsloth/DeepSeek-R1-Distill-Qwen-1.5B
language:
- en
license: apache-2.0
library_name: transformers
tags:
- deepseek
- qwen
- qwen2
- unsloth
- transformers
- TensorBlock
- GGUF
---

<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>

[![Website](https://img.shields.io/badge/Website-tensorblock.co-blue?logo=google-chrome&logoColor=white)](https://tensorblock.co)
[![Twitter](https://img.shields.io/twitter/follow/tensorblock_aoi?style=social)](https://twitter.com/tensorblock_aoi)
[![Discord](https://img.shields.io/badge/Discord-Join%20Us-5865F2?logo=discord&logoColor=white)](https://discord.gg/Ej5NmeHFf2)
[![GitHub](https://img.shields.io/badge/GitHub-TensorBlock-black?logo=github&logoColor=white)](https://github.com/TensorBlock)
[![Telegram](https://img.shields.io/badge/Telegram-Group-blue?logo=telegram)](https://t.me/TensorBlock)


## unsloth/DeepSeek-R1-Distill-Qwen-1.5B - GGUF

This repo contains GGUF format model files for [unsloth/DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/unsloth/DeepSeek-R1-Distill-Qwen-1.5B).

The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4823](https://github.com/ggml-org/llama.cpp/commit/5bbe6a9fe9a8796a9389c85accec89dbc4d91e39).

## Our projects
<table border="1" cellspacing="0" cellpadding="10">
<tr>
  <th style="font-size: 25px;">Awesome MCP Servers</th>
  <th style="font-size: 25px;">TensorBlock Studio</th>
</tr>
  <tr>
    <th><img src="https://imgur.com/2Xov7B7.jpeg" alt="Project A" width="450"/></th>
    <th><img src="https://imgur.com/pJcmF5u.jpeg" alt="Project B" width="450"/></th>
  </tr>
  <tr>
    <th>A comprehensive collection of Model Context Protocol (MCP) servers.</th>
    <th>A lightweight, open, and extensible multi-LLM interaction studio.</th>
  </tr>
<tr>
  <th>
    <a href="https://github.com/TensorBlock/awesome-mcp-servers" target="_blank" style="
      display: inline-block;
      padding: 8px 16px;
      background-color: #FF7F50;
      color: white;
      text-decoration: none;
      border-radius: 6px;
      font-weight: bold;
      font-family: sans-serif;
    ">👀 See what we built 👀</a>
  </th>
  <th>
    <a href="https://github.com/TensorBlock/TensorBlock-Studio" target="_blank" style="
      display: inline-block;
      padding: 8px 16px;
      background-color: #FF7F50;
      color: white;
      text-decoration: none;
      border-radius: 6px;
      font-weight: bold;
      font-family: sans-serif;
    ">👀 See what we built 👀</a>
  </th>
</tr>
</table>
## Prompt template

```
<｜begin▁of▁sentence｜>{system_prompt}<｜User｜>{prompt}<｜Assistant｜><think>
```

## Model file specification

| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q2_K.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q2_K.gguf) | Q2_K | 0.753 GB | smallest, significant quality loss - not recommended for most purposes |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q3_K_S.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q3_K_S.gguf) | Q3_K_S | 0.861 GB | very small, high quality loss |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q3_K_M.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q3_K_M.gguf) | Q3_K_M | 0.924 GB | very small, high quality loss |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q3_K_L.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q3_K_L.gguf) | Q3_K_L | 0.980 GB | small, substantial quality loss |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q4_0.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q4_0.gguf) | Q4_0 | 1.066 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q4_K_S.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q4_K_S.gguf) | Q4_K_S | 1.072 GB | small, greater quality loss |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q4_K_M.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q4_K_M.gguf) | Q4_K_M | 1.117 GB | medium, balanced quality - recommended |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q5_0.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q5_0.gguf) | Q5_0 | 1.259 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q5_K_S.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q5_K_S.gguf) | Q5_K_S | 1.259 GB | large, low quality loss - recommended |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q5_K_M.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q5_K_M.gguf) | Q5_K_M | 1.285 GB | large, very low quality loss - recommended |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q6_K.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q6_K.gguf) | Q6_K | 1.464 GB | very large, extremely low quality loss |
| [DeepSeek-R1-Distill-Qwen-1.5B-Q8_0.gguf](https://huggingface.co/tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/blob/main/DeepSeek-R1-Distill-Qwen-1.5B-Q8_0.gguf) | Q8_0 | 1.895 GB | very large, extremely low quality loss - not recommended |


## Downloading instruction

### Command line

Firstly, install Huggingface Client

```shell
pip install -U "huggingface_hub[cli]"
```

Then, downoad the individual model file the a local directory

```shell
huggingface-cli download tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF --include "DeepSeek-R1-Distill-Qwen-1.5B-Q2_K.gguf" --local-dir MY_LOCAL_DIR
```

If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:

```shell
huggingface-cli download tensorblock/DeepSeek-R1-Distill-Qwen-1.5B-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
```