AI BriefingGoogle Open-Sources ML Drift for GPU-Accelerated On-Device AI Inference
Key point
The new Apache 2.0 licensed engine abstracts OpenGL ES, OpenCL, Metal, and WebGPU to enable real-time ML experiences across platforms.
Details
The Google AI Edge Team announced the open-source release of ML Drift, a high-performance, cross-platform GPU compute engine designed specifically for on-device AI/ML inference. Released under the Apache 2.0 license, the tool aims to simplify the development of real-time, interactive ML applications ranging from advanced video effects to generative AI.
Hardware Abstraction and Performance
ML Drift abstracts the complexities of low-level APIs and hardware differences across major graphics standards:
- OpenGL ES
- OpenCL
- Metal
- WebGPU
By providing a unified foundation, the engine allows developers to achieve peak performance across multiple platforms without managing individual hardware constraints.
Integration and Availability
ML Drift serves as the core GPU acceleration engine within LiteRT. However, it is also available as a standalone library, enabling integration into custom graphics and inference runtimes for broader use cases.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
bigarrow: macOS CLI Tool Lets AI Agents Draw Visual Pointers on Screen