AI News

Trending story

NVIDIA NIM Certified Microservices Switch to SGLang Inference Engine

1 reports1 story updatesUpdated Nov 13 08:00

Understand this story

AI overview

On November 13, 2026, the LMSYS Blog published an article stating that SGLang now powers NVIDIA NIM Certified microservices, and that enterprises can adopt SGLang's inference capabilities through validated containers, optimized configurations for supported models and GPUs, and NVIDIA AI Enterprise support. The article explains that NIM's validation covers more than 30 NVIDIA hardware SKUs and more than 37 popular models, and that workloads are tuned based on SGLang capabilities such as Unified Radix Cache and Breakable CUDA Graph. Enterprise support is optional and requires an active NVIDIA AI Enterprise subscription. There are currently no subsequent developments.

AI-generated from public reports

Latest development · Report

SGLang and NVIDIA NIM: Advancing Enterprise-Grade Inference Together

Developments

1 confirmed developments

  1. NVIDIA NIM certified microservices switch to the SGLang inference engine

    Source:LMSYS Blog

Reports and discussion timeline

Browse reports and discussion by time. Their evidence levels remain distinct.

Nov 13

  1. LMSYS BlogFirst-party

    SGLang and NVIDIA NIM: Advancing Enterprise-Grade Inference Together

    SGLang now powers NVIDIA NIM Certified microservices, enabling enterprises to adopt SGLang's inference capabilities through validated containers, optimized configurations for supported models and GPUs, and NVIDIA AI Enterprise support. NIM's validation covers more than 30 NVIDIA hardware SKUs and more than 37 popular models, with workloads tuned based on SGLang capabilities such as Unified Radix Cache and Breakable CUDA Graph. Enterprise support is optional and requires an active NVIDIA AI Enterprise subscription.

Related stories

  1. Related storiesBaidu Wenxin releases ERNIE 5.5 Flash Preview
  2. Related storiesOpenAI introduces security controls for OpenClaw persistent agents
  3. Related storiesMastra launches private model gateway
  4. Related storiesBlock develops open-source collaboration space Buzz