---
title: QwQ-32B-FP8
canonical_url: "https://www.modelscope.cn/models/okwinds/QwQ-32B-FP8"
md_url: "https://www.modelscope.cn/models/okwinds/QwQ-32B-FP8.md"
repository: okwinds/QwQ-32B-FP8
last_updated: 2025-04-17
license: apache-2.0
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - qwen2
architectures:
  - Qwen2ForCausalLM
base_model:
  - Qwen/QwQ-32B
base_model_relation: quantized
parameters: 32.8B
tensor_type:
  - BF16
  - F8_E4M3
  - F32
library_name:
  - safetensors
  - pytorch
frameworks:
  - Pytorch
language:
  - en
  - zh
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy 0.9.1"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 717
stars: 2
tags:
  - Instruct
  - Chat
  - "Reason Model"
  - Quantization
---

# QwQ-32B-FP8

> QwQ-32B-FP8 - okwinds 在 ModelScope 开源的模型。正式版 QwQ-32B 的量化版本已出，用了更优质的数据做校准。欢迎品尝：QwQ-32B-Int8-W8A16

okwinds/QwQ-32B-FP8 是 ModelScope 魔搭社区上的 32.8B 参数text-generation模型，采用 apache-2.0 许可，基于 Qwen/QwQ-32B 构建，可用 deploy_task text/emb、lmdeploy 0.9.1、lmdeploy_turbomind 0.9.1 部署。

- **Repository**: okwinds/QwQ-32B-FP8
- **License**: apache-2.0
- **Tasks**: text-generation
- **Parameters**: 32.8B
- **Base model**: Qwen/QwQ-32B
- **Inference backends**: deploy_task text/emb, lmdeploy 0.9.1, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Tags**: Instruct, Chat, Reason Model, Quantization
- **Downloads**: 717
- **Stars**: 2
- **Last updated**: 2025-04-17

Source: https://www.modelscope.cn/models/okwinds/QwQ-32B-FP8

---

<span style="color:red">正式版 QwQ-32B 的量化版本已出，用了更优质的数据做校准。欢迎品尝：[QwQ-32B-Int8-W8A16](https://www.modelscope.cn/models/okwinds/QwQ-32B-Int8-W8A16)</span>

<span style="color:red">友情提醒，推理本模型时，请按照如下官方指引：</span>

> 注意：建议在运行这些模型时设置适当的温度（0.5 到 0.7 之间），否则您可能会遇到无休止的重复或不连贯的输出问题。

📖 关于量化损失方面的研究，可阅读公众号“觉察流”文章👇</br>

《[Reason Model 的“瘦身计划”：量化技术的得与失](https://mp.weixin.qq.com/s/NMGq4UUkfo8GMix5LHnWCg)》

# 通义千问 QwQ-32B-FP8

原模型 [Qwen/QwQ-32B](https://www.modelscope.cn/models/Qwen/QwQ-32B)

---

#### _作者在此 👇🏻 扫一扫_

<img src="https://www.modelscope.cn/models/okwinds/GPT-2/resolve/master/qrcode_for_jcl_258.jpg" />

---

## 下载

SDK下载
```bash
#安装ModelScope
pip install modelscope
```
```python
#SDK模型下载
from modelscope import snapshot_download
model_dir = snapshot_download('okwinds/QwQ-32B-FP8')
```
Git下载
```
#Git模型下载
git clone https://www.modelscope.cn/okwinds/QwQ-32B-FP8.git
```

## 模型概述

QwQ-32B-FP8 是一个基于 通义千问 QwQ-32B 的 FP8 量化模型。

- **模型名称:** QwQ-32B-FP8
- **模型架构:** Qwen2.5
- **权重量化:** FP8

该模型通过将 [qwen/QwQ-32B](https://www.modelscope.cn/models/Qwen/QwQ-32B) 的权重量化为 FP8 数据类型而实现。

量化过程将每个参数从 16bit 减少到 8bit，将模型占用磁盘空间大小，以及推理时加载模型需要的GPU显存空间，减少到了大约为原模型的1/2。

FP8（Floating Point 8-bit）是一种8位浮点数数据格式，由Nvidia、Arm、Intel联合推出，旨在加速深度学习的训练和推理过程。
本模型的量化，采用 F8_E4M3 数据类型对权重进行 dynamic 量化，尽可能减少模型量化后所带来的精度损失。

> <span style="color: red;">注意：本模型需要 compute capability > 8.0（Ampere、Ada Lovelace、Hopper 架构）的 Nvidia GPU 来支持 FP8 **混合精度**计算。</span>

## 部署推理

#### 推荐使用 vLLM

Openai api 兼容模式

```bash
>>> vllm serve "/home/gavin/llm/QwQ-32B-FP8" --host 0.0.0.0 --port 8000 --gpu-memory-utilization 0.9 --served-model-name "QwQ-32B-FP8"
```

---

# 附录 QwQ-32B 介绍

<a href="https://chat.qwenlm.ai/" target="_blank" style="margin: 2px;">
    <img alt="Chat" src="https://img.shields.io/badge/%F0%9F%92%9C%EF%B8%8F%20Qwen%20Chat%20-536af5" style="display: inline-block; vertical-align: middle;"/>
</a>

## Qwen2.5 同系列模型

#### 同系量化

| **Model**    | GPTQ Compression & FP8                                                                                                                                                                                                                                                                                                                                                                                                     | GGUF                                                                                         | **MLX** (macOS)                                                                                                                                                                                                                                                              |
| ------------ |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| -------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| *3B params*  | -                                                                                                                                                                                                                                                                                                                                                                                                                          | [3B-GGUF-V3-LOT](https://www.modelscope.cn/models/okwinds/Qwen2.5-3B-Instruct-GGUF-V3-LOT)   | [3B-MLX-4bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-3B-Instruct-MLX-4bit)<br/>[3B-MLX-8bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-3B-Instruct-MLX-8bit)                                                                                                |
| *7B params*  | [7B-Int4-W4A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-7B-Instruct-Int4-W4A16)<br/>[7B-Int8-W8A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-7B-Instruct-Int8-W8A16)<br/>[7B-FP8-A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-7B-Instruct-FP8-A16) <br/>[Coder-7B-Instruct-FP8](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-7B-Instruct-FP8) | [7B-GGUF-V3-LOT](https://www.modelscope.cn/models/okwinds/Qwen2.5-7B-Instruct-GGUF-V3-LOT)   | [7B-MLX-4bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-7B-Instruct-MLX-4bit)<br/>[7B-MLX-8bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-7B-Instruct-MLX-8bit)                                                                                                |
| *Coder 7B params* | [Coder-7B-Int4-W4A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-7B-Instruct-Int4-W4A16)<br/>[Coder-7B-Int8-W8A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-7B-Instruct-Int8-W8A16) <br/>[Coder-7B-FP8-A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-7B-Instruct-FP8-A16)       | [Coder-7B-GGUF-V3-LOT](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-7B-Instruct-GGUF-V3-LOT) | [Coder-7B-MLX-4bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-7B-Instruct-MLX-4bit)<br/>[Coder-7B-MLX-8bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-7B-Instruct-MLX-8bit)   |
| *14B params* | [14B-Int4-W4A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-14B-Instruct-Int4-W4A16)<br/>[14B-Int8-W8A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-14B-Instruct-Int8-W8A16)<br/>[14B-FP8-A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-14B-Instruct-FP8-A16)                                                                                                                                         | [14B-GGUF-V3-LOT](https://www.modelscope.cn/models/okwinds/Qwen2.5-14B-Instruct-GGUF-V3-LOT) | [14B-MLX-4bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-14B-Instruct-MLX-4bit)<br/>[14B-MLX-8bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-14B-Instruct-MLX-8bit)                                                                                            |
| *Coder 14B params* | [Coder-14B-Int4-W4A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-14B-Instruct-Int4-W4A16)<br/>[Coder-14B-Int8-W8A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-14B-Instruct-Int8-W8A16)                                                                                                                                        | [Coder-14B-GGUF-V3-LOT](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-14B-Instruct-GGUF-V3-LOT) | [Coder-14B-MLX-4bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-14B-Instruct-MLX-4bit)<br/>[Coder-14B-MLX-8bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-14B-Instruct-MLX-8bit)  |
| *32B params* | [32B-Int4-W4A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-32B-Instruct-Int4-W4A16)<br/>[32B-Int8-W8A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-32B-Instruct-Int8-W8A16)<br/>[32B-FP8-A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-32B-Instruct-FP8-A16)                                                                                                                                                                                                                                        | [32B-GGUF-V3-LOT](https://www.modelscope.cn/models/okwinds/Qwen2.5-32B-Instruct-GGUF-V3-LOT) | [32B-MLX-2bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-32B-Instruct-MLX-2bit)<br/>[32B-MLX-4bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-32B-Instruct-MLX-4bit)<br/>[32B-MLX-8bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-32B-Instruct-MLX-8bit) |
| *Coder 32B params* | [Coder-32B-Int4-W4A16](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-32B-Instruct-Int4-W4A16)<br/>[Coder-32B-FP8](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-32B-Instruct-FP8)                                                                                                                                                                                                                                        | [Coder-32B-GGUF-V3-LOT](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-32B-Instruct-GGUF-V3-LOT) | [Coder-32B-MLX-4bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-32B-Instruct-MLX-4bit)<br/>[Coder-32B-MLX-8bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-Coder-32B-Instruct-MLX-8bit) |
| *QwQ-Preview 32B params* | [QwQ-32B-Preview-Int4-W4A16](https://www.modelscope.cn/models/okwinds/QwQ-32B-Preview-Int4-W4A16)<br/>[QwQ-32B-Preview-FP8](https://www.modelscope.cn/models/okwinds/QwQ-32B-Preview-FP8)    | [QwQ-32B-GGUF-V3-LOT](https://www.modelscope.cn/models/okwinds/QwQ-32B-Preview-GGUF-V3-LOT) | [QwQ-32B-MLX-4bit](https://www.modelscope.cn/models/okwinds/QwQ-32B-Preview-MLX-4bit)<br/>[QwQ-32B-MLX-8bit](https://www.modelscope.cn/models/okwinds/QwQ-32B-Preview-MLX-8bit) |
| *72B params* | -                                                                                                                                                                                                                                                                                                                                                                                                                          | [72B-GGUF-V3-LOT](https://www.modelscope.cn/models/okwinds/Qwen2.5-72B-Instruct-GGUF-V3-LOT) | [72B-MLX-2bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-72B-Instruct-MLX-2bit)<br/>[72B-MLX-4bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-72B-Instruct-MLX-4bit)<br/>[72B-MLX-8bit](https://www.modelscope.cn/models/okwinds/Qwen2.5-72B-Instruct-MLX-8bit) |


## Introduction

QwQ is the reasoning model of the Qwen series. Compared with conventional instruction-tuned models, QwQ, which is capable of thinking and reasoning, can achieve significantly enhanced performance in downstream tasks, especially hard problems. QwQ-32B is the medium-sized reasoning model, which is capable of achieving competitive performance against state-of-the-art reasoning models, e.g., DeepSeek-R1, o1-mini.

<p align="center">
  <img width="100%" src="https://www.modelscope.cn/models/Qwen/QwQ-32B/resolve/master/figures/benchmark.jpg">
</p>


**This repo contains the QwQ 32B model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training (Supervised Finetuning and Reinforcement Learning)
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
- Number of Parameters: 32.5B
- Number of Paramaters (Non-Embedding): 31.0B
- Number of Layers: 64
- Number of Attention Heads (GQA): 40 for Q and 8 for KV
- Context Length: Full 131,072 tokens

**Note:** For the best experience, please review the [usage guidelines](#usage-guidelines) before deploying QwQ models.

You can try our [demo](https://huggingface.co/spaces/Qwen/QwQ-32B-Demo) or access QwQ models via [QwenChat](https://chat.qwen.ai).

For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwq-32b/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).

## Requirements

QwQ is based on Qwen2.5, whose code has been in the latest Hugging face `transformers`. We advise you to use the latest version of `transformers`.

With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```



## Usage Guidelines

To achieve optimal performance, we recommend the following settings:

1. **Enforce Thoughtful Output**: Ensure the model starts with "\<think\>\n" to prevent generating empty thinking content, which can degrade output quality. If you use `apply_chat_template` and set `add_generation_prompt=True`, this is already automatically implemented, but it may cause the response to lack the \<think\> tag at the beginning. This is normal behavior.

2. **Sampling Parameters**:
   - Use Temperature=0.6 and TopP=0.95 instead of Greedy decoding to avoid endless repetitions.
   - Use TopK between 20 and 40 to filter out rare token occurrences while maintaining the diversity of the generated output.

3. **No Thinking Content in History**: In multi-turn conversations, the historical model output should only include the final output part and does not need to include the thinking content. This feature is already implemented in `apply_chat_template`.

4. **Standardize Output Format**: We recommend using prompts to standardize model outputs when benchmarking.
   - **Math Problems**: Include "Please reason step by step, and put your final answer within \boxed{}." in the prompt.
   - **Multiple-Choice Questions**: Add the following JSON structure to the prompt to standardize responses: "Please show your choice in the `answer` field with only the choice letter, e.g.,`\"answer\": \"C\"`." in the prompt.

5. **Handle Long Inputs**: For inputs exceeding 32,768 tokens, enable [YaRN](https://arxiv.org/abs/2309.00071) to improve the model's ability to capture long-sequence information effectively.

For supported frameworks, you could add the following to `config.json` to enable YaRN:
```json
{
  ...,
  "rope_scaling": {
    "factor": 4.0,
    "original_max_position_embeddings": 32768,
    "type": "yarn"
  }
}
```

For deployment, we recommend using vLLM. Please refer to our [Documentation](https://qwen.readthedocs.io/en/latest/deployment/vllm.html) for usage if you are not familar with vLLM.
Presently, vLLM only supports static YARN, which means the scaling factor remains constant regardless of input length, **potentially impacting performance on shorter texts**. 
We advise adding the `rope_scaling` configuration only when processing long contexts is required.

## Evaluation & Performance

Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwq-32b/).

For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
