ModelShip
Hugging Face Hub connected
GitHub

Choose the right model.
Know what it takes to run.

Describe a workload or browse the Hub. ModelShip turns live model metadata into infrastructure requirements and production-ready release configuration.

LLMOps lifecycle

From model demand to reliable delivery.

Version 1 builds the deployment plan. Jenkins execution and Kubernetes delivery follow in the connected platform.

01

Discover

Match demand to model tasks.

Version 1
02

Assess

Verify metadata and compatibility.

Version 1
03

Size

Recommend compute and SLOs.

Version 1
04

Configure

Generate release configuration.

Version 1
05

Validate

Run Jenkins quality gates.

Roadmap
06

Deliver

Deploy and verify on Kubernetes.

Roadmap
Live Hugging Face catalog

Popular deployment candidates

Loading models…

Analyzing model metadata