TFLite
TensorFlow Lite, now known as LiteRT, is a lightweight framework and runtime for deploying and running machine-learning models on mobile, embedded, and edge devices.
Deep Learning with Python, Third Edition
A framework for efficient on-device TensorFlow inference on mobile and embedded hardware.
- How it is used
- Deploys TensorFlow inference on phones and embedded hardware when privacy, latency, or connectivity rules out remote serving.
- How deep it goes
- Defines the deployment role and supported hardware, but stops short of explaining TFLite runtime internals or conversion details.
- Examples
- Uses smartphone spam detection and factory bad-cookie detection to motivate local inference.
- What you can do after
- Operate on-device inference by exporting a Keras model and selecting TFLite for supported mobile or embedded targets.
Read how Deep Learning with Python, Third Edition explains TFLite →
Deep Learning with R, Third Edition
TensorFlow Lite is a framework for running machine-learning inference efficiently on mobile, embedded, and resource-constrained devices.
- How it is used
- Deploys machine-learning inference locally when privacy, latency, connectivity, or device constraints make remote serving unsuitable.
- How deep it goes
- Defines the deployment role and supported device categories but does not detail conversion, optimization, or runtime implementation.
- Examples
- Uses smartphone spam detection and factory cookie flagging to illustrate privacy, latency, and hardware deployment choices.
- What you can do after
- Deploys a saved Keras model for inference on supported mobile, embedded, ARM, Raspberry Pi, or microcontroller devices.
Read how Deep Learning with R, Third Edition explains TFLite →
Also covered in
What to know first
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
Related concepts
- TensorFlowuses · TensorFlow is an open-source software platform for numerical computation and machine learning, supporting the development, training, and execution of computational models.
- get_tensor_detailsuses · get_tensor_details is a TensorFlow Lite Interpreter method that returns metadata describing the tensors in a loaded model.
- TFLiteConverterused by · TFLiteConverter is the TensorFlow Lite API that converts TensorFlow or SavedModel graphs into deployable TFLite models, including optional optimization and quantization.
- LiteRT-LMcontrasts with · LiteRT-LM is a runtime and application interface for deploying language models with pre-quantized bundles and selectable execution backends.
- Optimumproduced by · Optimum is a Hugging Face library for optimizing and deploying machine-learning models across inference engines and hardware platforms.
- delegateuses · A delegate is a runtime extension that replaces or accelerates supported machine-learning operators with an alternate backend implementation.
Questions about TFLite
What is TFLite?
TensorFlow Lite, now known as LiteRT, is a lightweight framework and runtime for deploying and running machine-learning models on mobile, embedded, and edge devices.
Which Manning books explain TFLite?
Deep Learning with Python, Third Edition and Deep Learning with R, Third Edition.
What should I know before learning TFLite?
Manning books that teach TFLite usually introduce language model first.
What is the difference between TFLite and LiteRT-LM?
TensorFlow Lite, now known as LiteRT, is a lightweight framework and runtime for deploying and running machine-learning models on mobile, embedded, and edge devices. LiteRT-LM is a runtime and application interface for deploying language models with pre-quantized bundles and selectable execution backends.
What is TFLite also called?
TFLite is also called TensorFlow Lite.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

