Deploy AI across mobile, web, and embedded applications
-
On device
Reduce latency. Work offline. Keep your data local & private.
-
Cross-platform
Run the same model across Android, iOS, web, and embedded.
-
Multi-framework
Compatible with JAX, Keras, PyTorch, and TensorFlow models.
-
Full AI edge stack
Flexible frameworks, turnkey solutions, hardware accelerators
Ready-made solutions and flexible frameworks
Low-code APIs for common AI tasks
Cross-platform APIs to tackle common generative AI, vision, text, and audio tasks.
Get started with MediaPipe tasksDeploy custom models cross-platform
Performantly run JAX, Keras, PyTorch, and TensorFlow models on Android, iOS, web, and embedded devices, optimized for traditional ML and generative AI.
Get started with LiteRT
Shorten development cycles with visualization
Visualize your model’s transformation through conversion and quantization. Debug hotspots by overlaying benchmarks results.
Get started with Model Explorer
Build custom pipelines for complex ML features
Build your own task by performantly chaining multiple ML models along with pre and post processing logic. Run accelerated (GPU & NPU) pipelines without blocking on the CPU.
Get started with MediaPipe Framework
The tools and frameworks that power Google's apps
Explore the full AI edge stack, with products at every level — from low-code APIs down to hardware specific acceleration libraries.
MediaPipe Tasks
Quickly build AI features into mobile and web apps using low-code APIs for common tasks spanning generative AI, computer vision, text, and audio.
Generative AI
Integrate generative language and image models directly into your apps with ready-to-use APIs.
Vision
Explore a large range of vision tasks spanning segmentation, classification, detection, recognition, and body landmarks.
Text & audio
Classify text and audio across many categories including language, sentiment, and your own custom categories.
Get started
MediaPipe Framework
A low level framework used to build high performance accelerated ML pipelines, often including multiple ML models combined with pre and post processing.
LiteRT
Deploy AI models authored in any framework across mobile, web, and microcontrollers with optimized hardware specific acceleration.
Multi-framework
Convert models from JAX, Keras, PyTorch, and TensorFlow to run on the edge.
Cross-platform
Run the same exact model on Android, iOS, web, and microcontrollers with native SDKs.
Lightweight & fast
LiteRT’s efficient runtime takes up only a few megabytes and enables model acceleration across CPU, GPU, and NPUs.
Get started
Model Explorer
Visually explore, debug, and compare your models. Overlay performance benchmarks and numerics to pinpoint troublesome hotspots.
Gemini Nano in Android & Chrome
Build generative AI experiences using Google's most powerful, on-device model
