Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

Google Open-Sources ML Drift for Cross-Platform On-Device AI Inference

AI summary

Google released ML Drift under the Apache 2.0 license as an open-source, cross-platform GPU compute engine for on-device AI and machine-learning inference. It abstracts GPU hardware and low-level APIs across OpenGL ES, OpenCL, Metal, and WebGPU, powers LiteRT, and includes unified shaders, extensible custom operators, and native five-dimensional tensor support. Google says it is already running in Chrome, YouTube Shorts, Photos, Meet, Adobe apps, and Snapchat, with reported gains including lower memory overhead and faster inference or image-processing pipelines.

In-depth insight

ML Drift gives developers an Apache 2.0-licensed way to build portable inference and graphics runtimes across several GPU backends, while Google recommends migrating from the older TensorFlow Lite GPU delegate to the LiteRT ML Drift accelerator.

谷歌发布 ML Drift,一个开源、跨平台的 GPU 计算引擎,专为端侧 AI 和机器学习推理而设计"。它采用 Apache 2.0 许可,抽离了不同 GPU 硬件和底层 API 的复杂性,支持 OpenGL ES、OpenCL、Metal 和 WebGPU。

ML Drift 是为 LiteRT 提供动力的核心 GPU 加速引擎。谷歌将其开源后,开发者可以为各类应用构建自定义图形渲染与推理运行时,应用范围涵盖传统计算机视觉、音频和深度处理,以及生成式 AI 等场景。

ML Drift 的主要目标之一是简化开发者在创建设备端应用时面临的多样化 GPU 硬件的访问问题。

开发者面对着各种各样的 GPU 架构、驱动版本和底层 API,并且事先无法确定应用最终会在哪一种硬件上运行。

#Google#ML Drift#LiteRT#Dawn#OpenGL ES#OpenCL#Metal#WebGPU#Arm#Intel#Qualcomm#Gemma

Original · 3 min readRead the original
Group chat

No comments yet. Start the conversation.