NVIDIA/apex
apex
A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch
Builder
NVIDIA
NVIDIA • big-tech
Stars
8,939
Using upstream star count
Forks
1,518
Using upstream fork count
Open Issues
0
Activity Score
0/100
0 commits in 30d
Created
Apr 23, 2018
Project creation date
README Summary
NVIDIA Apex is a PyTorch extension that provides tools for easy mixed precision and distributed training. It offers automatic mixed precision (AMP) capabilities and optimized CUDA kernels to accelerate deep learning training. The library helps reduce memory usage and increase training speed while maintaining model accuracy.
AI Dev Skills
Unmapped
Mixed Precision TrainingDistributed Deep LearningGPU Memory OptimizationNeural Network Training AccelerationAutomatic Mixed Precision (AMP)Multi-GPU TrainingPyTorch Optimization
Tags
Mixed Precision TrainingDistributed Deep LearningGPU Memory OptimizationNeural Network Training AccelerationAutomatic Mixed Precision (AMP)Multi-GPU TrainingPyTorch OptimizationOn-premiseComputer Vision Model TrainingScalable Deep LearningMulti-GPU Model TrainingLarge Language Model TrainingImageTextMulti-GPU SystemsLarge Language ModelsEfficient AI TrainingDeep Neural Network OptimizationVideoMemory-Efficient TrainingCloudMultimodalAudioPython
Taxonomy
Deployment Context
Modalities
Skill Areas
Recent Activity
Updated 1 months ago
7 Days
0
30 Days
0
90 Days
0
Quality
production- Quality
- high
- Maturity
- production
Categories
Other AI / MLPrimaryMultimodal AIDev Tools & AutomationInference & ServingCoding & Dev ToolsFoundation ModelsModel TrainingGenerative MediaComputer Vision
PM Skills
Developer Platform
Languages
Python100.0%
Timeline
- Project created
- Apr 23, 2018
- Forked
- Mar 14, 2026
- Your last push
- 1 months ago
- Upstream last push
- 7 days ago
- Tracked since
- Mar 10, 2026
Similar Repos
pgvector cosine similarity · $0
Loading…