GitHub / NVIDIA / TransformerEngine
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference.
JSON API: http://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA%2FTransformerEngine
PURL: pkg:github/NVIDIA/TransformerEngine
Stars: 2,559
Forks: 459
Open issues: 266
License: apache-2.0
Language: Python
Size: 11 MB
Dependencies parsed at: Pending
Created at: almost 3 years ago
Updated at: 7 days ago
Pushed at: 7 days ago
Last synced at: 7 days ago
Topics: cuda, deep-learning, fp8, gpu, jax, machine-learning, python, pytorch