EdgeAI.computer

Chip manufacturer

Google

Publishes the Coral Edge TPU accelerators for low-power on-device inference and maintains LiteRT, the on-device runtime that succeeded TensorFlow Lite, together with the open-weight Gemma model family.

Visit website
Mountain View, California, United StatesProfile reviewed editorially

Categories

Edge AI Chips
Local AI Models
Developer Tools
On-Device AI

Industries served

Retail
Agriculture
Manufacturing

Solutions from Google

Small machine learning accelerator module beside a coin for scale
Google

Google Coral M.2 Accelerator (Edge TPU)

Low-power 4 TOPS Edge TPU module rated at 2 TOPS per watt for small, always-on inference workloads.

AI performance
4 TOPS (INT8) peak, 2 TOPS per watt
Interface
PCIe Gen2 x1 (M.2) or USB 2.0 / 3.0 (USB Accelerator)
Form factors
M.2 A+E, M.2 B+M, USB, solder-down module 15.0 × 10.0 × 1.5 mm
Deployment
Embedded board, single-board computer or small enclosure
Always-on sensor vision
Compact retail devices
Embedded classification
View DetailsCompareWebsite
Mobile device running an on-device machine learning feature
Google

Google LiteRT

Apache 2.0 on-device runtime for machine learning and generative AI, the successor to TensorFlow Lite.

Type
On-device machine learning and generative AI runtime
Licence
Apache 2.0
History
Renamed from TensorFlow Lite in September 2024
Deployment
Phone, embedded device or single-board computer
Mobile and appliance inference
Microcontroller and SBC vision
Offline features
View DetailsCompareWebsite

Company information is summarised from publicly available material. If you represent Google and would like to update or expand this profile, please get in touch.