Skip to main content

High-performance AI data loader written in Rust

Project description

Green AI 🚀

The World's Fastest Image Loader for PyTorch & TensorFlow.

PyPI version License: MIT

The Problem

Your GPUs are starving. Standard Python DataLoader pipelines are single-threaded and fight against the GIL (Global Interpreter Lock). While your H100 waits for data, you burn money.

The Solution

Green AI is a high-performance ingestion engine written in Rust. It bypasses Python's overhead to load, decode, and batch images in parallel, delivering them to your model via Zero-Copy memory transfer.

⚡ Performance Benchmark (1,000 Images)

Loader Time Speed Status
Standard Python (PIL) 17.48s 57 img/sec 🐢 Slow
Green AI (Rust) 1.22s 819 img/sec 🚀 14.3x Faster

(Benchmarks run on local dataset, 512x512 RGB JPEGs)

📦 Installation

pip install green-ai

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

turbo_pixel_loader-0.1.0-cp314-cp314-win_amd64.whl (854.1 kB view details)

Uploaded CPython 3.14Windows x86-64

File details

Details for the file turbo_pixel_loader-0.1.0-cp314-cp314-win_amd64.whl.

File metadata

File hashes

Hashes for turbo_pixel_loader-0.1.0-cp314-cp314-win_amd64.whl
Algorithm Hash digest
SHA256 40df36ef32381ef1a1be138d769e50f5d03ea2a0fc69d3d913f3b33b5e2deb34
MD5 f4bc3e04e5ffe8b4378f40b79769e51d
BLAKE2b-256 05971f2d7a7ef0b008a7a9ad58d81b5512a73821aaa4fd78089011a37ed95831

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page