DeploymentSpec

New files declare the stable v1 file contract:

apiVersion: infercrane.dev/v1
kind: Deployment

Pre-v1 files without the header are interpreted as v1. Unknown versions, kinds, and fields fail closed. See Upgrade and compatibility for the change policy. The primary path uses defaults:

infercrane deploy Qwen/Qwen3-8B

The YAML form groups model, runtime, compute, provider, resources, scaling, and routing concerns. Mutable model revisions such as main are resolved and persisted as immutable Hugging Face commits before provisioning.

name: qwen-prod
model:
  id: Qwen/Qwen3-8B
  revision: main
runtime:
  engine: vllm
  version: 0.10.2
  args:
    - --enable-prefix-caching
compute:
  mode: elastic
provider:
  cloud: runpod
  # adapter: skypilot # optional exact profile selection
  region: EU-RO-1
resources:
  gpu: L40S
scaling:
  min_replicas: 1
  max_replicas: 4
routing:
  strategy: round-robin

SGLang uses a built-in immutable workload profile. Custom OCI images declare the full contract in runtime.workload; see Custom OCI workloads. Mutable image tags and underspecified probes are rejected before provisioning. compute.mode defaults to elastic; elastic replica bounds default to 1..1. Serverless requires min_replicas: 0 and defaults max_replicas to one when omitted. Revisions are immutable; changing a field creates a candidate. provider.adapter persists the exact implementation when more than one profile can serve a cloud/runtime pair; it is optional for an unambiguous default. Current executable elastic defaults include configured AWS EC2 and GCP Compute, while other provider-product profiles remain registered but deferred until their exact combination is qualified. AWS and GCP require an explicit provider.region. Cost is omitted unless a trustworthy provider measurement exists.

GET STARTED

BUILD

OPERATE

INTELLIGENCE

INTEGRATE

REFERENCE

PROJECT

DeploymentSpec

DeploymentSpec

​DeploymentSpec

DeploymentSpec