Skip to content
Trending storyDeveloping

SGLang publishes Q4 multi-hardware roadmap covering 10+ accelerator families

1 article1 sourcesince Nov 11Last article

Overview

AISummary of 1 article

SGLang (@sgl_project) publishes its Q4 multi-hardware roadmap, saying the inference framework now runs on more than 10 accelerator families.

The roadmap lists in-tree Q4 focus areas for NVIDIA, AMD (MI45x, Helios, Ryzen AI Halo/Radeon), Google TPU, Ascend NPU, Intel Xeon CPU and XPU, T-Head PPU and Cambricon MLU.

AWS Trainium and MetaX are listed as coming through the plugin system rather than in-tree. SGLang says that plugin system makes adding out-of-tree backends a first-class path.

Written by AI from the articles below · updated Nov 11, 8:12 PM ET

Check the sources:

Latest developmentNov 11, 8:00 PM ET
SGLang publishes Q4 roadmap covering 10+ accelerator families

Article timeline

The articles in this story. Times are ET.

Nov 11
  1. SGLangOfficial
    SGLang publishes Q4 roadmap covering 10+ accelerator families

    AISGLang publishes its Q4 multi-hardware roadmap, saying it now runs on more than 10 accelerator families and that its plugin system makes adding out-of-tree backends a first-class path. The roadmap lists in-tree Q4 focus areas for NVIDIA, AMD (MI45x, Helios, Ryzen AI Halo/Radeon), Google TPU, Ascend NPU, Intel Xeon CPU and XPU, T-Head PPU and Cambricon MLU, with AWS Trainium and MetaX coming through the plugin.

    Image from @sgl_project's post

Heat Trend

Current comparable heat
7
Peak comparable heat
10Nov 11, 9:00 PM ET
Change in past 24 hours
–

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.