accelerate

A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support.

homepage: https://github.com/huggingface/accelerate

version	versionsuffix	toolchain
0.33.0	-CUDA-12.1.1	foss/2023a

(quick links: (all) - 0 - a - b - c - d - e - f - g - h - i - j - k - l - m - n - o - p - q - r - s - t - u - v - w - x - y - z)