# google-ai-edge/LiteRT

LiteRT, successor to TensorFlow Lite. is Google's On-device framework for high-performance ML & GenAI deployment on edge platforms, via efficient conversion, runtime, and optimization

Repository: https://github.com/google-ai-edge/LiteRT
Canonical: https://ross.abutalabs.com/products/litert
Homepage: https://ai.google.dev/edge/litert/next/overview
Language: C++
License: Apache-2.0
License Family: permissive
Last push: 2026-08-26T22:39:17+00:00

## Health v2 (maintenance only)
Score: 85/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 85, longevity 52
- inputs: {"age_days": 728, "days_push": 7, "days_rel": 20, "gap_med": 38, "n_releases_24m": 14}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3339, forks 434 (observed 2026-08-28T04:07:56.827293+00:00)

## What it is
LiteRT is Google's successor to TensorFlow Lite, an on-device runtime for high-performance ML and GenAI inference on edge platforms. It provides model conversion, CPU/GPU/NPU acceleration, and optimization tooling for deploying models on Android, iOS, embedded, and web targets.

## Use cases
- run machine learning models on android devices
- deploy llm inference on mobile phones
- accelerate on-device inference with gpu and npu
- convert tensorflow models for edge deployment
- run ml models in the browser with webgpu
- deploy genai models on embedded hardware

## When to choose
- you need fast on-device ML inference on Android, iOS, or embedded platforms
- you want to run LLMs or GenAI models locally without a server
- you need hardware acceleration across CPU, GPU, and NPU from one runtime
- you are migrating an existing TensorFlow Lite deployment

## When to avoid
- you need server-side or cloud GPU training rather than edge inference
- you need a full training framework rather than a deployment runtime
- your target platform lacks LiteRT delegate support for your accelerator

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, llm-inference, gpu-computing, sdk, cli
- domain: machine-learning, deep-learning, large-language-models, mobile-development, embedded-systems, developer-tools
- platform: windows, cpp, python, wasm, cross-platform, embedded
- tags: tensorflow-lite, on-device-ai, edge-ai, inference-runtime, genai, npu-acceleration, model-conversion, android, ios, linux, macos

## Member repositories
- google-ai-edge/LiteRT (main) score 85

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:56.827293+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:41:05.388081+00:00, confidence not recorded.
  - readme: https://github.com/google-ai-edge/LiteRT (fetched 2026-08-28T04:07:56.827293+00:00, sha 7edd3630c94c)
  - homepage: https://ai.google.dev/edge/litert/next/overview (fetched 2026-08-29T09:34:17.782137+00:00, sha 861f450bd8d2)
- Data as of 2026-08-30T08:39:29.467469+00:00.