Multihull

Deploy hot GPU containers to many providers. One URL. Automatic failover.

Quickstart Read the design

One spec

Describe image, GPU class, scaling and targets once in multihull.yaml. Each provider gets its native payload.

Always warm

At least two providers stay hot by default. Fallbacks never scale to zero. Health beats price.

Per-request failover

A stateless Rust router retries across providers before the first byte is committed. Callers see one URL.

A minimal spec

apiVersion: multihull/v1
name: llama-8b

container:
  image: ghcr.io/acme/vllm-llama:1.4.0
  port: 8000
  health: {path: /health}

resources:
  gpu: [L4, A10G]

targets:
  - provider: gke-prod
    type: kubernetes
    priority: 1
  - provider: modal-main
    type: modal
    priority: 2

route:
  hostname: llama.api.acme.com
  protocol: openai

What `hull status` shows

Target	State	Replicas	GPU
gke-prod	Ready	2/2	L4
modal-main	Degraded	1/1	A10G
runpod-eu	Serving failover	1/1	L4

Start with the Quickstart, or read how targets and failover work.