# openxla/xla

A machine learning compiler for GPUs, CPUs, and ML accelerators

Repository: https://github.com/openxla/xla
Canonical: https://ross.abutalabs.com/products/xla
Language: C++
License: Apache-2.0
License Family: permissive
Last push: 2026-08-26T22:58:22+00:00

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 100
- inputs: {"age_days": 1485, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4499, forks 911 (observed 2026-08-28T04:08:51.451871+00:00)

## What it is
XLA (Accelerated Linear Algebra) is an open-source machine learning compiler that optimizes models from frameworks like PyTorch, TensorFlow, and JAX for high-performance execution. It targets GPUs, CPUs, and ML accelerators across different hardware platforms.

## Use cases
- compile PyTorch models for faster GPU execution
- optimize TensorFlow models for ML accelerators
- speed up JAX computations on CPUs and GPUs
- build custom hardware backends for ML frameworks
- debug and integrate new ML frontends with the compiler

## When to choose
- you need to optimize ML models across diverse hardware backends
- you are developing or integrating a new ML accelerator
- you use PyTorch, TensorFlow, or JAX and want compiler-driven performance gains

## When to avoid
- you just want to use XLA from a framework without contributing to the compiler itself
- you need a simple inference runtime rather than a compiler toolchain

## Facets
- artifact type: library
- maturity: active
- function: compiler, machine-learning, gpu-computing, llm-training
- domain: machine-learning, compilers, gpu-computing, deep-learning
- platform: cpp, cross-platform
- tags: xla, ml-compiler, linear-algebra, accelerators, openxla, linux, gpu

## Member repositories
- openxla/xla (main) score 77

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:51.451871+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:20:26.575921+00:00, confidence not recorded.
  - readme: https://github.com/openxla/xla (fetched 2026-08-28T04:08:51.451871+00:00, sha 98fc97dc1b47)
- Data as of 2026-08-30T08:39:29.467469+00:00.