Tesseract
Get started
Tesseract Core Tesseract-JAX Tesseract-Torch
Demos Blog User Forum
Tesseract Core Tesseract-JAX Tesseract-Torch
Star Star
Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
Tesseract Core 1.15.1
Light Logo Dark Logo

Introduction

  • Overview
  • Installation
  • Changelog

Tutorials & examples

  • Get started
  • Creating Tesseracts
  • Interacting with Tesseracts
  • Demos & tutorials
    • Inverse heat transfer with automatic differentiation
    • Multi-physics pipeline with end-to-end gradient optimization
    • Learned closure: Training a neural viscosity model through a PDE solver (PyTorch)
    • Parametric shape optimization with differentiable FEM simulation
    • 4D-variational data assimilation for a chaotic dynamical system
    • Building the JAX solver Tesseract for Lorenz-96
    • Simulator-in-the-loop Bayesian inference with Tesseract
    • Gradient-based optimization of fluid flows
    • JAX Rosenbrock Minimization
    • PyTorch Rosenbrock Minimization
    • JAX RBF Fitting
  • Example gallery: Building blocks
    • HelloWorld
    • VectorAdd
    • Univariate Rosenbrock function
    • Wrapping compiled code (Fortran example)
    • Differentiating compiled code (Enzyme + Fortran)
    • Wrapping MATLAB code
    • Packaging local files into a Tesseract
    • Custom build steps: PyVista on ARM64
    • Installing local Python modules into a Tesseract
    • Installing dependencies from a lockfile
    • Out-of-core data loading
    • File IO with InputPath / OutputPath
    • Returning on-disk arrays with BinrefArray
    • Finite difference gradients
    • Deriving gradient endpoints from each other
  • Example gallery: Integrating with Ansys products
    • Wrapping SpaceClaim as a Tesseract
    • Wrapping MAPDL as a Tesseract

How-to guides

  • Tips for defining Tesseract APIs
  • Composing Tesseracts into pipelines
  • Advanced usage
  • Deploying Tesseracts
  • Checking gradients
  • Fast same-machine runs with shared-memory binref
  • Keep GPU arrays on the device with cuda_ipc
  • Debugging guide

Concepts

  • When to use Tesseract
  • Design patterns
  • Differentiable programming basics
  • Performance trade-offs & optimization

Reference β€” SDK

  • CLI reference (tesseract)
  • Python API (tesseract_core)
  • Configuration (tesseract_config.yaml)

Reference β€” runtime

  • Endpoints (tesseract_api.py)
  • Array encodings reference
  • CLI reference (tesseract-runtime)
  • Python API (tesseract_core.runtime)
Back to top
View this page

Example gallery: Integrating with Ansys productsΒΆ

This is a collection of examples where Tesseract is integrated with Ansys products in various ways.

The Ansys software suite is heavily used throughout the engineering simulation industry. Wrapping Ansys products as Tesseracts allows you to embed them into Tesseract-driven compute pipelines β€” acting as powerful data generators, differentiable solvers, geometry engines, and more.

See also

You can find the code for all Ansys Tesseracts in the demo/_showcase directory of the Tesseract Core repository.

SpaceClaim

A Tesseract that wraps SpaceClaim for CAD geometry creation.

Wrapping SpaceClaim as a Tesseract
MAPDL

A differentiable Tesseract that wraps the MAPDL solver via PyMAPDL with an analytic adjoint, for use in SIMP topology optimization.

Wrapping MAPDL as a Tesseract
Next
Wrapping SpaceClaim as a Tesseract
Previous
Deriving gradient endpoints from each other
Copyright © 2026, Pasteur Labs
Made with Sphinx and @pradyunsg's Furo