TNN
A high-performance neural network inference framework supporting various deep learning frameworks and hardware platforms.
TNN: developed by Tencent Youtu Lab and Guangying Lab, a uniform deep learning inference framework for mobile、desktop and server. TNN is distinguished by several outstanding features, including its cross-platform capability, high performance, model compression and code pruning. Based on ncnn and Rapidnet, TNN further strengthens the support and performance optimization for mobile devices, and also draws on the advantages of good extensibility and high performance from existed open source efforts. TNN has been deployed in multiple Apps from Tencent, such as Mobile QQ, Weishi, Pitu, etc. Contributions are welcome to work in collaborative with us and make TNN a better framework.
Related projects:
| Repository | Description | Stars |
|---|---|---|
| tencent/ncnn | An optimized framework for deploying deep learning models on mobile devices. | 20,479 |
 alibaba/mnn | A lightweight deep learning framework designed for efficient inference and training on-device | 8,739 |
| tencent/hunyuandit | A PyTorch-based diffusion transformer model for generating images with fine-grained Chinese understanding and text-to-image synthesis | 3,456 |
| tencent/pocketflow | A framework that automatically compresses and accelerates deep learning models to make them suitable for mobile devices with limited computational resources. | 2,788 |
 tflearn/tflearn | A high-level API for deep learning that builds upon TensorFlow | 9,619 |
 tusen-ai/simpledet | A framework for object detection and instance recognition with high-performance training capabilities | 3,084 |
 moli232777144/mtcnn_ncnn | A project that builds and integrates the ncnn framework for mobile face detection on Windows and Android platforms. | 528 |
 majianjia/nnom | A high-level inference Neural Network library specifically designed for microcontrollers. | 950 |
 sjtu-ipads/powerinfer | An efficient Large Language Model inference engine leveraging consumer-grade GPUs on PCs | 7,964 |
 nihui/ncnn-android-styletransfer | An Android project demonstrating style transfer using the NCNN neural network library | 90 |
 microsoft/cntk | A unified deep learning toolkit that describes neural networks as a series of computational steps via a directed graph. | 17,523 |
| tencent/tencent-hunyuan-large | This project makes a large language model accessible for research and development | 1,114 |
 xilinx/finn | Fast and scalable neural network inference framework for FPGAs. | 747 |
| microsoft/mmdnn | A toolset to convert and manage deep learning models across multiple frameworks. | 5,797 |
 alexlenail/nn-svg | A tool that generates publication-ready diagrams of neural network architectures using parametric methods and popular JavaScript libraries. | 5,088 |