---
license: other
license_name: gml
license_link: https://github.com/OpenBMB/General-Model-License/blob/main/%E9%80%9A%E7%94%A8%E6%A8%A1%E5%9E%8B%E8%AE%B8%E5%8F%AF%E5%8D%8F%E8%AE%AE-%E6%9D%A5%E6%BA%90%E8%AF%B4%E6%98%8E-%E5%AE%A3%E4%BC%A0%E9%99%90%E5%88%B6-%E5%95%86%E4%B8%9A%E6%8E%88%E6%9D%83.md
language:
- en
- zh
tags:
- MiniCPM
- ModelBest
- THUNLP
- TensorBlock
- GGUF
base_model: openbmb/MiniCPM-2B-sft-fp32
---

<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>
<div style="display: flex; justify-content: space-between; width: 100%;">
    <div style="display: flex; flex-direction: column; align-items: flex-start;">
        <p style="margin-top: 0.5em; margin-bottom: 0em;">
            Feedback and support: TensorBlock's  <a href="https://x.com/tensorblock_aoi">Twitter/X</a>, <a href="https://t.me/TensorBlock">Telegram Group</a> and <a href="https://x.com/tensorblock_aoi">Discord server</a>
        </p>
    </div>
</div>

## openbmb/MiniCPM-2B-sft-fp32 - GGUF

This repo contains GGUF format model files for [openbmb/MiniCPM-2B-sft-fp32](https://huggingface.co/openbmb/MiniCPM-2B-sft-fp32).

The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4242](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).

<div style="text-align: left; margin: 20px 0;">
    <a href="https://tensorblock.co/waitlist/client" style="display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">
        Run them on the TensorBlock client using your local machine ↗
    </a>
</div>

## Prompt template

```
{system_prompt}<用户>{prompt}<AI>
```

## Model file specification

| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [MiniCPM-2B-sft-fp32-Q2_K.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q2_K.gguf) | Q2_K | 1.204 GB | smallest, significant quality loss - not recommended for most purposes |
| [MiniCPM-2B-sft-fp32-Q3_K_S.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q3_K_S.gguf) | Q3_K_S | 1.355 GB | very small, high quality loss |
| [MiniCPM-2B-sft-fp32-Q3_K_M.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q3_K_M.gguf) | Q3_K_M | 1.481 GB | very small, high quality loss |
| [MiniCPM-2B-sft-fp32-Q3_K_L.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q3_K_L.gguf) | Q3_K_L | 1.564 GB | small, substantial quality loss |
| [MiniCPM-2B-sft-fp32-Q4_0.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q4_0.gguf) | Q4_0 | 1.609 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [MiniCPM-2B-sft-fp32-Q4_K_S.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q4_K_S.gguf) | Q4_K_S | 1.682 GB | small, greater quality loss |
| [MiniCPM-2B-sft-fp32-Q4_K_M.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q4_K_M.gguf) | Q4_K_M | 1.802 GB | medium, balanced quality - recommended |
| [MiniCPM-2B-sft-fp32-Q5_0.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q5_0.gguf) | Q5_0 | 1.914 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [MiniCPM-2B-sft-fp32-Q5_K_S.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q5_K_S.gguf) | Q5_K_S | 1.948 GB | large, low quality loss - recommended |
| [MiniCPM-2B-sft-fp32-Q5_K_M.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q5_K_M.gguf) | Q5_K_M | 2.045 GB | large, very low quality loss - recommended |
| [MiniCPM-2B-sft-fp32-Q6_K.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q6_K.gguf) | Q6_K | 2.367 GB | very large, extremely low quality loss |
| [MiniCPM-2B-sft-fp32-Q8_0.gguf](https://huggingface.co/tensorblock/MiniCPM-2B-sft-fp32-GGUF/blob/main/MiniCPM-2B-sft-fp32-Q8_0.gguf) | Q8_0 | 2.899 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/MiniCPM-2B-sft-fp32-GGUF --include "MiniCPM-2B-sft-fp32-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/MiniCPM-2B-sft-fp32-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
```