@jiayq Insane hardware & inference engineering! > 600 tok/s for GLM5.2 on Blackwell nodes is absolutely mind-blowing. The Fleet redefines autonomous infra building.
We are starting Intent Lab, building an autonomous team we call "fleet" that turns intent into production software. Today we are sharing some early results: the fastest GLM5.2 inference engine, one shot database creation, and a fully verified agent filesystem.
https://t.co/CtwtypE0rH
Learn how to run distributed deep learning training jobs on Amazon EKS using Kubeflow & the AWS FSx CSI driver, & how to optimize training performance to improve throughput & minimize training times. Jiaxin Shan, Software Engineer for Amazon EKS, explains. https://t.co/mV7R1186H3
Create #MachineLearning model -> deploy as an endpoint -> batch transform using validation datasets, all using #KubeFlow pipelines and Amazon Sagemaker:
https://t.co/hLOftBWNRQ
Thanks @ilovpitt!