Tencent/ncnn

ncnn

ncnn is a high-performance neural network inference framework optimized for mobile, embedded, and desktop deployment. It has no third-party runtime dependencies, runs across CPU and Vulkan GPU backends, and provides tools such as pnnx for converting PyTorch and ONNX models to ncnn. Developers can deploy deep learning models efficiently on phones, PCs, browsers, and edge devices. ncnn is currently being used in many Tencent applications, such as QQ, Qzone, WeChat, Pitu, and so on.

ncnn 是一个面向移动端、嵌入式和桌面端部署优化的高性能神经网络推理框架。 ncnn 无第三方运行时依赖，支持 CPU 和 Vulkan GPU 后端，并提供 pnnx 等工具将 PyTorch 和 ONNX 模型转换为 ncnn 模型。 基于 ncnn，开发者可以将深度学习模型高效部署到手机、PC、浏览器和边缘设备上。 ncnn 目前已在腾讯多款应用中使用，如：QQ，Qzone，微信，天天 P 图等。

---

Quick Start

The recommended beginner path is PyTorch -> pnnx -> ncnn.

pip3 install pnnx

import torch
import torch.nn as nn
import pnnx
class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d(3, 8, 1)
        self.relu = nn.ReLU()
        self.fc = nn.Linear(8, 4)
def forward(self, x):
        x = self.conv(x)
        x = self.relu(x)
        x = x.mean((2, 3))
        return self.fc(x)
model = Model().eval()
x = torch.rand(1, 3, 224, 224)
pnnx.export(model, "model.pt", (x,))

#include "net.h"
ncnn::Net net;
net.load_param("model.ncnn.param");
net.load_model("model.ncnn.bin");
ncnn::Mat in(224, 224, 3);
auto ex = net.create_extractor();
ex.input("in0", in);
ncnn::Mat out;
ex.extract("out0", out);

import numpy as np
import ncnn
net = ncnn.Net()
net.load_param("model.ncnn.param")
net.load_model("model.ncnn.bin")
x = np.zeros((3, 224, 224), np.float32)
mat = ncnn.Mat(x)
ex = net.create_extractor()
ex.input("in0", mat)
ret, out = ex.extract("out0")
print(np.array(out).shape)

See pnnx, use ncnn with PyTorch or ONNX, Python API, and examples for complete workflows.

---

Community

---

Download & Build status

https://github.com/Tencent/ncnn/releases/latest

	how to build ncnn library on Linux / Windows / macOS / Raspberry Pi3, Pi4 / POWER / Android / NVIDIA Jetson / iOS / WebAssembly / AllWinner D1 / Loongson 2K1000
Source
	Build for Android Build for Termux on Android
Android
Android shared
	Build for HarmonyOS with cross-compiling
HarmonyOS
HarmonyOS shared
	Build for iOS on macOS with xcode
iOS
iOS-Simulator
	Build for macOS
macOS
Mac-Catalyst
watchOS
watchOS-Simulator
tvOS
tvOS-Simulator
visionOS
visionOS-Simulator
Apple xcframework
	Build for Linux / NVIDIA Jetson / Raspberry Pi3, Pi4 / POWER
Ubuntu 22.04
Ubuntu 24.04
	Build for Windows x64 using VS2017 Build for Windows x64 using MinGW-w64
VS2015
VS2017
VS2019
VS2022
	Build for WebAssembly
WebAssembly
	Build for ARM Cortex-A family with cross-compiling Build for Hisilicon platform with cross-compiling Build for AllWinner D1 Build for Loongson 2K1000 Build for QNX
Linux (arm)
Linux (aarch64)
Linux (mips)
Linux (mips64)
Linux (ppc64)
Linux (riscv64)
Linux (loongarch64)

---

Build

Use the prebuilt packages above when possible. To build from source, see the full how to build ncnn library guide for Linux, Windows, macOS, Android, iOS, WebAssembly, HarmonyOS, Raspberry Pi, Jetson, and embedded targets.

Common Linux build:

git clone --recursive https://github.com/Tencent/ncnn.git
cd ncnn
mkdir build && cd build
cmake -DCMAKE_BUILD_TYPE=Release -DNCNN_VULKAN=ON -DNCNN_BUILD_EXAMPLES=ON ..
cmake --build . -j$(nproc)

---

Model Conversion

| Source model | Recommended path | Docs | | --- | --- | --- | | PyTorch | pnnx.export(model, "model.pt", (input_tensor,)) or pnnx model.pt inputshape=[...] | pnnx, PyTorch / ONNX guide | | ONNX | pnnx model.onnx | pnnx, onnx tools | | ncnn model optimization | ncnnoptimize model.param model.bin new.param new.bin flag | quantization, model file spec | | Legacy Caffe / MXNet / Darknet | Use compatibility converters when maintaining older models | caffe, mxnet, darknet, AlexNet legacy tutorial |

Use Netron to inspect .param, .onnx, and .pnnx.param graphs.

---

Features

• No third-party runtime dependencies and no BLAS / NNPACK requirement.
• Pure C++ implementation with C API and Python binding.
• Optimized CPU inference for mobile and embedded processors, including ARM NEON and multi-core scheduling.
• Vulkan GPU acceleration for supported platforms.
• Low memory footprint with explicit blob/workspace allocator design.
• Supports multi-input, multi-output, and multi-branch graphs.
• PyTorch and ONNX conversion through pnnx, plus legacy converter support for older model formats.
• Supports fp16 storage/arithmetic paths, int8 quantized inference, model optimization, and custom layers.
• Direct memory reference loading for .param and .bin models.

Model and Workload Coverage

ncnn is still strong for classic and mobile CNN workloads, but current usage is broader than CNN-only deployment.

• Classification and backbones: VGG, AlexNet, GoogleNet, Inception, ResNet, DenseNet, SENet, SqueezeNet, MobileNet, ShuffleNet, MNasNet.
• Detection and face: SSD, Faster R-CNN, R-FCN, MTCNN, RetinaFace, scrfd, YOLOv2, YOLOv3, YOLOv4, YOLOv5, YOLOv7, YOLOv8, YOLOX, NanoDet.
• Segmentation, pose, and OCR: FCN, PSPNet, UNet, YOLACT, SimplePose, PP-OCR examples.
• Audio, generation, and language workloads are represented by community projects and examples where the model operators are supported.

---

Project Examples

| Area | Project | | --- | --- | | Image generation | zimage-ncnn-vulkan - Z-Image g