proxy.golang.org : github.com/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.
Registry
-
Source
- Documentation
- JSON
- codemeta.json
purl: pkg:golang/github.com/nvidia/%21transformer%21engine
Keywords:
cuda
, deep-learning
, fp8
, gpu
, jax
, machine-learning
, python
, pytorch
License: Apache-2.0
Latest release: 8 months ago
First release: almost 2 years ago
Stars: 2,840 on GitHub
Forks: 528 on GitHub
See more repository details: repos.ecosyste.ms
Last synced: 30 days ago