← back to the directory
Library / SDK Fine-tuning & Training

Accelerate

1Run 2PyTorch 3training 4on 5any 6hardware.

the six Ws · specification

W1 Who

Maintained by Hugging Face.

W2 What

Abstracts device placement and distributed training behind a thin PyTorch wrapper.

W3 Where

A Python library on CPU, GPU, or TPU.

W4 When

When scaling a training loop across devices.

W5 Why

The same code runs on any setup.

W6 With

PyTorch.

W7 Watch

Official Hugging Face library, Apache-2.0; no credentials, runs locally.

Hugging-FaceDistributedPyTorch

for agents & scripts

Reading this as a machine? Query it directly.

Search is open JSON - no key. Report telemetry after using a tool and it feeds that tool’s Proof Score. Or speak MCP to /mcp and discover tools mid-loop.