OpenXLA project

license

Apache-2.0(OSI)

source

public

core features withheld

no

The LICENSE file contains the Apache-2.0 text. The repository is public and the README describes no paid tier, enterprise edition or license-gated build, so the core is ungated.

• https://raw.githubusercontent.com/openxla/xla/main/docs/build_from_source.md recorded 2026-09-30

  Build guide says "This document describes how to build XLA components." and builds from the repository with "./configure.py --backend=CUDA" for GPU. Searched the body for enterprise, license key, paid, proprietary and commercial: none.

• https://raw.githubusercontent.com/openxla/xla/main/LICENSE recorded 2026-09-30

  LICENSE file is the verbatim Apache-2.0 text, opening with "Apache License" and "Version 2.0, January 2004" and "TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION".

• https://raw.githubusercontent.com/openxla/xla/main/README.md recorded 2026-09-30

  README says "XLA (Accelerated Linear Algebra) is an open-source machine learning (ML)" compiler and that "If you're not contributing code to the XLA compiler, you don't need to clone and" build this repo, so the repository holds the compiler itself. Searched the body for enterprise, license key, paid, commercial, proprietary and closed source: none.

XLA ships inside the frameworks that embed it - PyTorch, TensorFlow and JAX - rather than as its own package, so GitHub stars are the only available adoption signal. Because XLA's actual reach is that of the frameworks built on it, the star count only partially reflects real usage.

• https://api.github.com/repos/openxla/xla recorded 2026-08-18

  Repo metadata - stargazers_count = 4,477 - for openxla/xla.

XLA is a retargetable, end-to-end compiler - an HLO optimization pipeline fed by PyTorch, TensorFlow and JAX front ends, emitting code for CPUs, GPUs and ML accelerators. That breadth of inputs and targets is comparable to Apache TVM's.

• https://raw.githubusercontent.com/openxla/xla/main/docs/architecture.md recorded 2026-09-30

  Architecture page says "The CPU and GPU backends included with XLA use [LLVM](http://llvm.org) for low-level IR," and "The GPU backend currently supports NVIDIA GPUs via the LLVM NVPTX backend. The CPU" backend supports multiple CPU ISAs; the frontends are "PyTorch, TensorFlow, and JAX".

• https://raw.githubusercontent.com/openxla/xla/main/docs/index.md recorded 2026-09-30

  Docs home says XLA "supports various backends including GPUs, CPUs, and ML" accelerators and "includes a pluggable infrastructure to add support for" more.

• https://raw.githubusercontent.com/openxla/xla/main/README.md recorded 2026-09-30

  README says "The XLA compiler takes models from popular ML frameworks such as PyTorch," TensorFlow, and JAX and optimizes them for "high-performance execution across" different hardware platforms including GPUs, CPUs, and ML accelerators.