MNN
Alibaba's lightweight deep learning engine for on-device inference and training.
About
MNN (Mobile Neural Network) is Alibaba's open-source inference and training engine optimized for mobile, embedded, server, and PC deployment. It is production-grade software integrated into more than 30 Alibaba apps including Taobao, Tmall, Youku, and DingTalk, covering dozens of scenarios from live broadcast to image search.
The framework supports TensorFlow, Caffe, and ONNX models, runs on CPU and GPU backends including Metal, OpenCL, and Vulkan, and is published under Apache 2.0. Its MNN-LLM and MNN-Diffusion runtimes extend it to on-device large language models and stable diffusion generation.
What it can do
- Lightweight inference with no third-party dependencies
- MNN-LLM runtime for on-device language models
- MNN-Diffusion runtime for on-device image generation
- CPU, Metal, OpenCL, and Vulkan backends
Setup
Clone the alibaba/MNN repository and build with CMake for your target platform, or use the prebuilt Android/iOS packages.
Prerequisites
CMake and a C++ toolchain for the target device; model files converted to MNN format.
Use cases
Works with
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 116; Trending score 74/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.