TransformerEngine
by NVIDIA (organization)
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference. (description from the project’s GitHub page)
Repository facts
- Stars
- 3,585
- Forks
- 866
- Open issues and pull requests
- 373
- Primary language
- Python
- License
- Apache License 2.0 (Apache-2.0)
- Owner
- NVIDIA · Organization
- Created
- Sep 20, 2022
- Last push
- Oct 16, 2026
Topics
Related projects
Projects that share this one’s GitHub topics.
Rank 6,523: mosecorg/mosec
A high-performance ML model serving framework, offers dynamic batching and CPU/GPU pipelines to fully exploit your compute machine
902 starsPythonApache-2.0Last push Oct 12, 2026
Rank 3,574: deepmodeling/deepmd-kit
A deep learning package for many-body potential energy representation and molecular dynamics
2,061 starsPythonLGPL-3.0Last push Oct 18, 2026
Rank 1,373: chainer/chainer
A flexible framework of neural networks for deep learning
5,920 starsPythonMITLast push Aug 28, 2023
Rank 6,965: jasmcaus/caer
High-performance Vision library in Python. Scale your research, not boilerplate.
820 starsPythonMITLast push Sep 23, 2026
Rank 3,339: kubeflow/trainer
Distributed AI Model Training and LLM Fine-Tuning on Kubernetes
2,249 starsGoApache-2.0Last push Oct 16, 2026
Rank 582: isl-org/Open3D
Open3D: A Modern Library for 3D Data Processing
14,102 starsC++Other licenseLast push Sep 30, 2026