Downloads · 30 days
31
1% of all-time downloads
QwQ-32B-Preview-MNN
taobao-mnn/QwQ-32B-Preview-MNN
QwQ-32B-Preview-MNN is a text generation model from taobao-mnn. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
This model is a 4-bit quantized version of the MNN model exported from QwQ-32B-Preview using llmexport.
All-time downloads
5.2K
Public
Repo size
19.6 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.weight18 GB · 92%
At a glance
- Task
- Text Generation
- License
- apache-2.0
- Access
- Public
- Created
- Nov 29, 2024
- Updated
- Dec 31, 2024
- SHA
- edec600a
Try a prompt
Model card
From the Hugging Face model README
QwQ-32B-Preview-MNN
Introduction
This model is a 4-bit quantized version of the MNN model exported from QwQ-32B-Preview using llmexport.
Download
# install huggingface
pip install huggingface
# shell download
huggingface download --model 'taobao-mnn/QwQ-32B-Preview-MNN' --local_dir 'path/to/dir'
# SDK download
from huggingface_hub import snapshot_download
model_dir = snapshot_download('taobao-mnn/QwQ-32B-Preview-MNN')
# git clone
git clone https://www.modelscope.cn/taobao-mnn/QwQ-32B-Preview-MNN
Usage
# clone MNN source
git clone https://github.com/alibaba/MNN.git
# compile
cd MNN
mkdir build && cd build
cmake .. -DMNN_LOW_MEMORY=true -DMNN_CPU_WEIGHT_DEQUANT_GEMM=true -DMNN_BUILD_LLM=true -DMNN_SUPPORT_TRANSFORMER_FUSE=true
make -j
# run
./llm_demo /path/to/QwQ-32B-Preview-MNN/config.json prompt.txt
Document
- Task
- Text Generation
- License
- apache-2.0
- Languages
- en
- Created
- Nov 29, 2024
- Updated
- Dec 31, 2024