Optimum documentation
Overview
Overview
Nvidia
Intel
AWS Trainium/Inferentia
Google TPUs
ExecuTorch
ONNX
Furiosa
Exporters
Torch FX
LLM quantization
Utilities
You are viewing main version, which requires installation from source. If you'd like
regular pip install, checkout the latest stable version (v1.27.0).
Overview
🤗 Optimum handles the export of TensorFlow models to TFLite in the exporters.tflite module. In addition, models hosted on the Hugging Face Hub with PyTorch weights but having a TensorFlow implementation will also be supported in the export thanks to Transformers’ TFPreTrainedModel.from_pretrained() auto-conversion to TensorFlow.
The TFLite export support provides classes, functions and a command line interface to export a model easily.
Supported architectures:
- Albert
- BERT
- Camembert
- ConvBert
- Deberta
- Deberta V2
- DistilBert
- Electra
- Flaubert
- MobileBert
- MPNet
- ResNet
- Roberta
- RoFormer
- XLM
- XLMRoberta