Google along with other AI/ML leaders including AMD, Arm, Google, Intel, Meta, NVIDIA, Apple, AWS and others today launched OpenXLA, an open-source ML compiler ecosystem.
OpenXLA will enable efficient lowering, optimization and deployment of ML models from most major frameworks such as PyTorch and TensorFlow to any hardware backend notably CPUs, GPUs, and ML ASICs without compromising on high performance.
OpenXLA will start with the XLA compiler, which is being decoupled from TensorFlow, and StableHLO, a portable ML compute operation set that makes frameworks easier to deploy across different hardware options.
OpenXLA is a community-led and open-source ecosystem of ML compiler and infrastructure projects
OpenXLA project has the following goals:
- Accelerate industry collaboration around XLA and build a vibrant OSS community.
- Share and receive feedback on the technical direction for OpenXLA and ensure it meets the needs of major users and contributors.
- Set up a new XLA repository or organization with independent build/test, with infra to more easily accept PRs, and that is hardware and framework independent.
- Ensure the extraction of XLA from TensorFlow is minimally disruptive to existing users and contributors.
- Create a product identity with its own brand, website, docs, and communication channels.
- Discuss establishment of governance outside TensorFlow.