Hugging Face Accelerate
Library for running PyTorch training across distributed setups with minimal code changes.
About
Accelerate by Hugging Face lets you run a plain PyTorch training loop across any hardware setup, from a single CPU to multi-GPU and TPU configurations with mixed precision, by adding only a few lines of code. It abstracts the boilerplate for distributed training and device placement while leaving the rest of the script unchanged, and supports fp16, bf16, and fp8. Released under the Apache 2.0 license.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- AI Frameworks & Libraries
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Easy (2/5)
- License
- Apache-2.0
- Minimum VRAM
- 4 GB
- Added
- Apr 3, 2026
Related Tools
Unified system for large-scale distributed training and inference.
Guidance language for controlling LLM generation with constraints, grammars, and JSON schemas.
Library for easily accessing and processing ML datasets.
JAX neural network library with PyTorch-like module system.
High-level deep learning API supporting JAX, TensorFlow, and PyTorch backends.
Gradient boosting library by Yandex with native categorical feature handling.