Catalog
Blueprint by nvidia
Serve LLMs with vLLM on One Device
A locally accessible model API from one DGX Spark or Station
NVIDIA blueprintvLLMInference128 GBDGX SparkDGX StationRun LLM Inference
DashboardManage jobs, terminals, files and account settings.DocsBuild against ICPX APIs, program IDL and payment flows.BrandingUse the ICPX name, logo and visual system correctly.
Compute
Compute on demandRun agents, models and ML workloads through ICPX.Model indexBrowse models, skills and blueprints.ModelsDeployable model catalog and runtime options.SkillsAgent-ready capabilities and tool patterns.BlueprintsOpen workload patterns for compute deployment.ProcessSee how a compute run is routed, priced and executed.
Build
Account
Company
Network