SGLang Partners with Google Cloud to Bring Full TPU Support for Inference Acceleration
hsu_byron · x · 2026-07-31
RadixArk and Google Cloud are collaborating with the SGLang community to make TPU a highly efficient path for frontier model inference. SGL-JAX already supports running major open-source model families on the latest TPU generations, including Gemma, Qwen, DeepSeek, and Grok, as well as video and image generation models.
This partnership integrates SGLang's core production features into the TPU ecosystem, including 5D parallelism, Radix Cache, quantization, and speculative decoding. Developers can leverage a native PyTorch path via SGL-torchtpu, ensuring a seamless inference experience on TPUs with the same API and feature set.
Related event: SGLang Joins Google Cloud and RadixArk to Bring TPU Inference(6 posts)→
More from Infra
- Meta's commitments jumped to $349B in a quarter as server memory prices tripled ahead of Oct 21 earnings — tengyanAI · 2026-10-16
- Meta releases Muse Spark Open, trained with vLLM-powered RL inference from day one — colinmcnamara · 2026-10-16
- Meta's open Muse Spark needs 493GB of memory — far beyond any consumer laptop — TejasKumar_ · 2026-10-16
- CoreWeave plunges 20% as credit markets price >60% default odds; is this dot-com all over again? — kimmonismus · 2026-10-16
- MiniCPM-V 4.7 35B-A3B hits 17+ tokens/s on iPad with ~3GB memory — techNmak · 2026-10-16
- China rolls out 'token factories' and a 2026-28 token economy action plan — nordicinst · 2026-10-16