Taichi Lang: High-performance Parallel Programming in Python

notification iconnotification iconTaichi NeRF is released, try it now!

Taichi Logo

Github stars

-:-

Get StartedUse CasesCommunityBlogDiscord

Taichi Logo

Github stars

-:-

Taichi Logo

Get Started

Use Cases

Community

Blog

High-performance

parallel programming

in Python

Taichi is a domain-specific language embedded in Python that helps you easily write portable, high-performance parallel programs.

Install

$ pip install taichi -UCopy Logo

Examples

$ ti galleryCopy Logo

Get Started

What’s possible with Taichi

Speed up PythonSpeed up Python

image

Taichi translates the compute-intensive Python code into fast machine code at runtime with a just-in-time compiler, accelerating your Python code.

Try now

Numerical simulationNumerical simulation

image

Create an urban airflow simulation using 3D Lattice Boltzmann Method, with spatially sparse data structures provided by Taichi.

Learn more

3D rendering3D rendering

image

A GPU path tracing voxel renderer made by Taichi! You can easily create inspiring voxel art using this renderer within 99 lines of Taichi code.

Learn more

Computer visionComputer vision

With automatic parallelization and spatially sparse data structures, Taichi powers the fast development of a LiDAR-based SLAM system.

Learn more

More Use Cases

The advantages of using Taichi

Easy to learn

No barrier to entry for Python users: Taichi shares almost the same syntax as Python. Apply a single Taichi decorator, and your functions are automatically turned into optimized machine code.

Easy to learn

Try drawing a Mandelbrot set in 40 lines of Taichi code

Easy to learn

High-performance

Taichi’s JIT compiler automatically compiles Python functions into fast GPU or CPU machine code for parallel execution. While Taichi lives in Python, it can approach or even outrun the speed of C++ or CUDA.

High-performance

View Taichi benchmarks

High-performance

Integrated into the Python ecosystem

Taichi can seamlessly interoperate with popular Python frameworks, such as NumPy, PyTorch, matplotlib, and pillow.

Integrated into the Python ecosystem

Learn how Taichi interacts with arrays from external libraries

Integrated into the Python ecosystem

Universal deployment

Taichi’s ahead-of-time mode enables deployment on platforms without Python, including PCs, mobile devices, and web browsers.

Universal deployment

Learn more about Taichi's AOT solution

Key features of Taichi

High-performance parallel computingHigh-performance parallel computing

A one-billion-particle MPM simulation running on a GPU with 80 GB memory - the world’s largest-scale single-GPU particle simulation!

Learn more

Automatic differentiationAutomatic differentiation

Soft robot locomotion control powered by Taichi’s handy automatic differentiation system, which is significantly faster than PyTorch, TensorFlow, and JAX.

Learn more

Flexible SNode systemFlexible SNode system

image

Taichi’s SNode system allows quick experimentation with different memory layouts to maximize performance without rewriting computation code.

Learn more

Spatially sparse data structuresSpatially sparse data structures

A fluid simulation example where empty regions take no computation nor memory, enabled by spatially sparse data structures in Taichi.

Learn more

User Page

Organizations that use Taichi

user logo

https://d33wubrfki0l68.cloudfront.net/cb20380683fa4a94f486c7596aff1df9180114d2/0d974/user-logo/card-eth.png

ETH Zürich uses Taichi in its Physically-based Simulation course, effectively improving both the learning and teaching experience.

View course

https://d33wubrfki0l68.cloudfront.net/ead2e6dd35a617fcaf2f24bb4173ba0b4f2ec49a/5308c/user-logo/utah.png

The University of Utah uses Taichi for simulation in the paper A General Two-Stage Initialization for Sag-Free Deformable Simulations, published at SIGGRAPH 2022.

View paper

https://d33wubrfki0l68.cloudfront.net/98b1f429be951a0da75802a5bb8b0d5214aea927/a5b4d/user-logo/oppo-en.png

OPPO, a smart device manufacturer, uses Taichi to develop stunning physics-based wallpapers that run interactively on mobile devices.

Learn more

https://d33wubrfki0l68.cloudfront.net/8b95b3b86a6dcaabbbc1ad21872f3b4a9503c396/dbe3b/user-logo/kuaishou-en.png

Kuaishou, a short video platform, uses Taichi as its high-performance mobile GPU solution, enabling unique real-time liquid and gas simulations for generating special effects.

What users love about Taichi

Taichi is mainly used to replace some of our GPU shaders code and make our development so much easier (especially for debugging!). In addition, Taichi supports generation of AOT modules, allowing us to run our code on different platforms including Android on Mobile phones! Creating beautiful and physical-based algorithms has never been so easy with Taichi!

Gabriel Huau

OPPO Senior Engineering Manager

Taichi is much easier to grasp than traditional GPU programming languages. Once a Taichi program is compiled, it can be deployed on iOS, Android, PC, and many more platforms. We need to iterate and release new (GPU-based) visual effects quickly, and Taichi has greatly accelerated our workflow.

Chongyang Ma

Lead of Y-tech Graphics AI, Kwai

The simplicity of design in Taichi is always the key to great work. With much less code, we achieved surprisingly good visual results and highly sophisticated features. Taichi proves to be a significant boost to my research and development work.

Hao Xu

Ph.D. student in Computer Engineering, HKUST

We use Taichi for physical simulation. It is easy to use. 90% of our projects used Taichi and were completed with flying colors. Taichi is also efficient. It has much less scaffolding code than other languages or frameworks.

Jingwei Tang

Research Assistant, ETH Zürich

User Page

Join the Taichi Community

together with over 200 contributors from all over the world

DiscordDiscord

Discord

Join our Discord channel to meet the Taichi developers and peer users/contributors, and chat about anything related to Taichi – be it Taichi's internals, RFCs, or new features!

GitHubGitHub

GitHub

You are more than welcome to file bug reports, open pull requests, track issues and milestones on GitHub.

TwitterTwitter

Twitter

YouTubeYouTube

YouTube

Subscribe to Taichi’s YouTube channel for useful tutorials and impressive demos.

Ready to dive in Taichi?

Install and write your first Taichi program!

Install Taichi

$ pip install taichi -UCopy Logo

Example demos

$ ti galleryCopy Logo

Get Started

Documentation

Cheatsheet

Tutorial

API Reference

FAQ

Resources

Blog

Newsletter

Release Notes

Global Forum

Community

Contributor's Guide

Good First Issue

Issue Tracker

Code of Conduct

Newsletter

Get the latest news from Taichi in a monthly email.

Taichi Logo

| Privacy Policy

User Page