DALI

A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.

View on GitHub↗Upstream NVIDIA/DALI↗

Builder

NVIDIA

NVIDIA • big-tech

Stars

5,659

Using upstream star count

Forks

660

Using upstream fork count

Open Issues

Activity Score

0/100

0 commits in 30d

Created

Jun 1, 2018

Project creation date

README Summary

NVIDIA DALI is a GPU-accelerated library designed to optimize data processing pipelines for deep learning applications. It provides highly optimized building blocks and an execution engine that accelerates data loading and preprocessing for both training and inference workloads. The library aims to remove data processing bottlenecks that can limit GPU utilization in machine learning workflows.

AI Dev Skills

Unmapped

Tags

GPU Computing and CUDA ProgrammingData Pipeline OptimizationDeep Learning InfrastructureComputer Vision PreprocessingHigh-Performance ComputingMemory Management and OptimizationParallel Data ProcessingNeural Network Training AccelerationAudio Preprocessing for Neural NetworksCloud GPU InstancesReal-time Data Pipeline AccelerationEfficient Neural Network TrainingAudioOn-premise GPU ClustersImageGPU-Accelerated ComputingSelf-hostedInference Pipeline OptimizationHigh-Performance ML InfrastructureLarge-scale Training Data ProcessingImage Data Augmentation for Computer VisionVideoVideo Processing for Deep LearningC++

Recent Activity

Updated 28 days ago

7 Days

30 Days

90 Days

Quality

production

Quality: high
Maturity: production

PM Skills

Scale & ReliabilityDeveloper Platform

Languages

C++100.0%

Timeline

Project created: Jun 1, 2018
Forked: Mar 14, 2026
Your last push: 28 days ago
Upstream last push: 7 days ago
Tracked since: Mar 17, 2026

Similar Repos

pgvector cosine similarity · $0

Loading…