Google Releases LiteRT Tensor API for Native Edge Model Authoring
Google announced the LiteRT Tensor API, a C++ authoring layer that lets developers construct computation graphs operation by operator for on-device AI. Google says the same graph can run through LiteRT on CPUs, ML Drift-powered GPUs and NPUs, or directly through XNNPACK on CPU, with reference implementations including SAM 2 and Gemma models.
The release gives AI developers a portable native-authoring option for tightly controlling edge inference graphs across CPU, GPU and NPU backends without maintaining separate implementations.
Native Model Authoring for Edge with LiteRT Tensor API - Google Developers Blog
Native Model Authoring for Edge with LiteRT Tensor API
OCT. 19, 2026
Ping Yu Staff Software Engineer
Quentin Khan Sr. Software Engineer
Chintan Parikh Sr. Product Manager
#Google#LiteRT Tensor API#LiteRT#ML Drift#XNNPACK#SAM 2#Gemma 4#Gemma 3