---
title: How-tos
description: >-
  - Convert to Fabric in 5 minutes — Learn how to add Fabric to your PyTorch
  code - Scale your model with Accelerators — Take advantage of your hardware...
canonical: 'https://lightning.ai/docs/fabric/stable/more/guide'
---
# How-tos {#how-tos}

## Basics {#basics}

- [Convert to Fabric in 5 minutes](/docs/fabric/2.6.6/home/installation/convert.md) — Learn how to add Fabric to your PyTorch code

- [Scale your model with Accelerators](/docs/fabric/2.6.6/home/installation/convert/accelerators.md) — Take advantage of your hardware with a switch of a flag

- [Structure your Fabric code](/docs/fabric/2.6.6/more/code_structure.md) — Best practices for setting up your training script with Fabric

- [Launch distributed training](/docs/fabric/2.6.6/reference/fundamentals/launch.md) — Launch a Python script on multiple devices and machines

- [Launch Fabric in a notebook](/docs/fabric/2.6.6/reference/fundamentals/notebooks.md) — Launch on multiple devices from within a Jupyter notebook

- [Improve performance with Mixed-Precision training](/docs/fabric/2.6.6/reference/fundamentals/precision.md) — Save memory and speed up training using mixed precision

## Build your own Trainer {#build-your-own-trainer}

- [Organize your model code with LightningModule](/docs/fabric/2.6.6/reference/guide/lightning_module.md) — Organize your code in a LightningModule and use it with Fabric

- [Encapsulate code into Callbacks](/docs/fabric/2.6.6/reference/guide/callbacks.md) — Make use of the Callback system in Fabric

- [Track and visualize experiments](/docs/fabric/2.6.6/reference/guide/logging.md) — Learn how Fabric helps you remove boilerplate code for tracking metrics with a logger

- [Save and load model progress](/docs/fabric/2.6.6/reference/guide/checkpoint.md) — Efficient saving and loading of model weights, training state, hyperparameters and more.

- [Build your own Trainer](https://github.com/Lightning-AI/lightning/tree/master/examples/fabric/build_your_own_trainer) — Take our Fabric Trainer template and customize it for your needs

## Advanced Topics {#advanced-topics}

- [Use efficient gradient accumulation](/docs/fabric/2.6.6/reference/advanced/gradient_accumulation.md) — Learn how to perform efficient gradient accumulation in distributed settings

- [Distribute communication](/docs/fabric/2.6.6/reference/advanced/distributed_communication.md) — Learn all about communication primitives for distributed operation. Gather, reduce, broadcast, etc.

- [Use multiple models and optimizers](/docs/fabric/2.6.6/reference/advanced/multiple_setup.md) — See how flexible Fabric is to work with multiple models and optimizers!

- [Speed up models by compiling them](/docs/fabric/2.6.6/reference/advanced/compile.md) — Use torch.compile to speed up models on modern hardware

- [Train models with billions of parameters](/docs/fabric/2.6.6/reference/advanced/model_parallel.md) — Train the largest models with FSDP/TP across multiple GPUs and machines

- [Initialize models efficiently](/docs/fabric/2.6.6/reference/advanced/model_init.md) — Reduce the time and peak memory usage for model initialization

- [Save and load very large models](/docs/fabric/2.6.6/reference/guide/checkpoint/distributed_checkpoint.md) — Save and load very large models efficiently with distributed checkpoints

---

**Navigation**

- Parent: [More](/docs/fabric/2.6.6/more.md)