HIP-410: LLM CRD. Status Active. Hanzo's own standard — read this before implementing against it.
The LLM CRD is the facade Kind for the Hanzo LLM gateway (hanzoai/gateway, which historically also lived as hanzoai/llm). It proxies 100+ upstream model providers and exposes a unified OpenAI-compatible API. The reconciler delegates to the Service controller. See HIP-004 and HIP-044 for design.
hanzo.ai/v1, plural llms, shortname llm.
Same shape as Service (HIP-400). Conventionally:
image.repository: ghcr.io/hanzoai/gatewayports: containerPort: 4000gateway-secrets via envFrom (ANTHROPIC_API_KEY, OPENAI_API_KEY, etc.)SQL, KV, and consoleapiVersion: hanzo.ai/v1
kind: LLM
metadata:
name: gateway
namespace: hanzo
spec:
image:
repository: ghcr.io/hanzoai/gateway
tag: 2.14.1
replicas: 2
ports:
- name: http
containerPort: 4000
servicePort: 80
envFrom:
- secretRef:
name: gateway-secrets
resources:
requests: { cpu: 500m, memory: 1Gi }
limits: { cpu: 2, memory: 4Gi }
Deployment, Service, optional HPA.
~/work/hanzo/operator/src/controllers/service.rs (via LLM facade).
Implemented in hanzoai/operator v0.3.0+. CR gateway active in cluster.