基于 PyTorch 的目标检测全家桶，覆盖训练、推理、实例分割、图像分类和模型导出，开箱即用，适合快速落地视觉任务。
把 ONNX、PyTorch、Keras、CoreML 等格式的神经网络模型直接拖进浏览器就能看结构图，排查模型转换和层连接问题很方便。
….1](https://github.com/mrousavy/react-native-fast-tflite/compare/v3.0.0...v3.0.1) (2026-04-21) Bug Fixes return created GPU delegate from getAndroidGPUDelegate ([ 184](ht
….0](https://github.com/mrousavy/react-native-fast-tflite/compare/v3.0.0-beta.0...v3.0.0) (2026-04-14) In react-native-fast-tflite v3.0.0 we migrated the entire module to
….0](https://github.com/mrousavy/react-native-fast-tflite/compare/v2.0.0...v3.0.0-beta.0) (2026-04-14) Bug Fixes support Bridgeless mode using RCTTurboModuleWithJSIBinding
….0](https://github.com/mrousavy/react-native-fast-tflite/compare/v1.6.1...v2.0.0) (2026-01-13) Bug Fixes Add 16KB page alignment and upgrade LiteRT to 1.4.0 ([ 163](https
….1](https://github.com/mrousavy/react-native-fast-tflite/compare/v1.6.0...v1.6.1) (2025-04-08)
…/github.com/Ar-Ray-code/YOLOX-ROS/pull/26 support tflite C++ by @fateshelled in https://github.com/Ar-Ray-code/YOLOX-ROS/pull/31 Update package.xml by @Ar-Ray-code in htt