Taichi (太极) is a programming language designed for high-performance computer graphics. It is deeply embedded in Python, and its just-in-time compiler offloads compute-intensive tasks to multi-core CPUs and massively parallel GPUs.
Advanced features of Taichi include spatially sparse computing and differentiable programming [examples].
python3 -m pip install taichiSupported OS: Windows, Linux, Mac OS X; Python: 3.6, 3.7, 3.8; Backends: x64 CPUs, CUDA, Apple Metal, OpenGL Compute Shaders.
Please build from source for other configurations (e.g., your CPU is ARM).
Note:
- Starting April 13 2020 (v0.5.12), we release the Python package
taichiinstead oftaichi-nightly. Now this PyPI package includes CPU, CUDA 10, Metal and OpenGL support. - On Ubuntu 19.04+, please
sudo apt install libtinfo5. - On Windows, please install Microsoft Visual C++ Redistributable if you haven't.
- [All releases] [Change log]
| Linux (CUDA) | OS X (10.14+) | Windows | Documentation | |
|---|---|---|---|---|
| Build |  |
|||
| PyPI |  |
|
- A High-Performance Multi-Material Continuum Physics Engine based on Taichi: Taichi Elements (work in progress)
The Taichi project was created by Yuanming Hu (yuanming-hu). Significant contributions are made by:
- Ye Kuang (k-ye) (Apple Metal backend)
- 彭于斌 (archibate) (OpenGL Compute Shader backend)
- Mingkuan Xu (xumingkuan) (IR optimization & standardization)
Kenneth Lozes (KLozes) and Yu Fang (squarefk) have also made notable contributions. [List of all contributors]
We welcome feedback and comments. If you would like to contribute to Taichi, please check out our Contributor Guidelines.
If you use Taichi in your research, please cite our papers: