google-ai-edge /mediapipe

Cross-platform, customizable ML solutions for live and streaming media.

★ 37,706⑂ 6,228👁 530⑃ 651● 5,709▤ 72
C++Apache-2.0Created 2019-06-13ai.google.dev/edge/mediapipe
下载:MediaPipe v1.2.3MediaPipe-1.2.3 ↓1.2.32026-11-11
+17
日增长
+88
周增长
+408
月增长
37,706
stars
6,228
forks
530
watchers
651
prs
5,709
issues
72
releases

README

Attention: See https://developers.google.com/edge/mediapipe
for the primary developer documentation site for MediaPipe.

!MediaPipe

On-device machine learning for everyone

Delight your customers with innovative machine learning features. MediaPipe
contains everything that you need to customize and deploy to mobile (Android,
iOS), web, desktop, edge devices, and IoT, effortlessly.

   See demos
   Learn more

 Get started

You can get started with MediaPipe Solutions by following the developer guides
for each solution, including:

   audio
   decision
   genai
   retrieval
   text
   vision

If you need help setting up a development environment for use with MediaPipe
Tasks, follow our setup guides for:

   Android
   iOS
   Python
   Web apps

 Solutions

MediaPipe Solutions provides a suite of libraries and tools for you to quickly
apply artificial intelligence (AI) and machine learning (ML) techniques in your
applications. You can plug these solutions into your applications immediately,
customize them to your needs, and use them across multiple development
platforms. MediaPipe Solutions is part of the MediaPipe open source
project, so you can further
customize the solutions code to meet your application needs.

These libraries and resources provide the core functionality for each MediaPipe
Solution:

   MediaPipe Tasks: Cross-platform APIs and libraries for deploying
    solutions. Learn
    more.
   MediaPipe models: Pre-trained, ready-to-run models for use with each
    solution.

To customize and evaluate solutions, use MediaPipe Studio. It lets you
visualize, test, and benchmark solutions in your browser.
Learn more and
see our demos.

 Framework

To start using MediaPipe Framework, install MediaPipe
Framework
and start building example applications in C++, Android, and iOS.

MediaPipe Framework is
the low-level component used to build efficient on-device machine learning
pipelines, similar to the premade MediaPipe Solutions.

Before using MediaPipe Framework, familiarize yourself with the following key
Framework
concepts:

   Packets
   Graphs
   Calculators

 Community

   Discuss - General
    community discussion around MediaPipe.
   Awesome MediaPipe - A
    curated list of awesome MediaPipe related frameworks, libraries and
    software.

 Contributing

We welcome contributions. Please follow these
guidelines.

We use GitHub issues for tracking requests and bugs. Please post questions to
the MediaPipe Stack Overflow with a mediapipe tag.

 Privacy Notice

Last modified: June 5, 2026

When you use MediaPipe Tasks, processing of the input data (e.g. images, video,
text) takes place on device, and MediaPipe does not send that input data to
Google servers. As a result, you can use our MediaPipe Tasks APIs for processing
data that should not leave the device.

MediaPipe Tasks APIs send metrics about the performance and utilization of the
APIs in your app to Google. Google uses this metrics data to measure
performance, usage, debug, mainta

Top issues

Latest release

MediaPipe v1.2.3 v1.2.3 2026-11-11
Bazel changes

- Migrate mediapipe::NoDestructor to absl::NoDestructor (absl/base/no_destructor.h) in mediapipe/framework and remove obsolete mediapipe/framework/deps/no_destructor.h.
- Respect TensorBufferRequirements alignment in InferenceRunnerLiteRt
- Resolve file-path media parts in SemanticRetriever::Retrieve before embedding.
- [MediaPipe] Fix tensor view destruction deadlock on LiteRT run error.
- Register only the CPU accelerator in LiteRT environments for CPU-only runners.
- Add NPU dispatch library directory to C MpBaseOptions.
- Forward BaseOptions app telemetry in C++ task

Snapshot: 2026-11-16T09:23:14Z · Data license Apache-2.0 · JSON