Deployment

    One platform. Three ways to run it.

    The same governed AI coding platform, deployed where your data rules require it. Start on our cloud. Move inference into your network. Or run the whole thing inside your perimeter, air-gapped.

    Same policy engine in all threeAir-gapped is self-hosted, zero outbound networkStart SaaS. Migrate when you are ready.
    We run itSplit the stackYou run all of it
    Kimchi cloud
    Your network
    Cloud managed

    Kimchi Cloud

    We run the control plane, the gateway, and the models. You call an endpoint.

    Control planeLLM gatewayInference · our GPUs
    Your IDE & agents
    Prompts and code go to Kimchi
    Hybrid

    Your inference, our control plane

    The gateway and models live in your network. We run the console and governance.

    Control planeConsole
    LLM gatewayInference · your GPUsYour IDE & agents
    Minimal operational metadata
    Self-hosted

    The full stack, inside

    The whole platform in your cluster. One Helm chart. Air-gapped is this, with reporting off.

    Nothing runs on our side
    Control planeLLM gatewayConsoleDatabasesInference · your GPUsYour IDE & agents
    Optional reporting · off when air-gapped
    Hybrid · Managed inference

    Your execution, our models

    The gateway and remote workspaces run in your network. Inference is served by Kimchi.

    Inference · our GPUs
    LLM gatewayRemote workspacesYour IDE & agents
    Prompts from the workspaces go to Kimchi

    What runs where. What leaves your network.

    The same policy engine in every pattern. What changes is who operates it and which bytes cross the boundary.

    ComponentKimchi CloudHybridSelf-hosted
    Control planeKimchi cloudKimchi cloudYour cluster
    LLM gatewayKimchi cloudYour clusterYour cluster
    InferenceKimchi cloudOpen models on our GPUs.Your clusterYour models, or Kimchi serverless as overflow.Your clusterYour models. Any in-perimeter endpoint.
    ConsoleKimchi cloudKimchi cloudYour cluster
    Analytics storeKimchi cloudKimchi cloudYour cluster
    IdentityKimchi cloudKimchi cloudYour IdPOIDC. Nothing to provision on our side.
    Leaves your networkPrompts and codeMinimal operational metadataNothing, with reporting disabledOptional reporting, off in the air-gapped configuration.

    Runs inside. Answers to your team.

    Self-hosted puts the whole platform on your infrastructure, under your identity, behind your monitoring. No dependency on a round trip to ours.

    Your Kubernetes cluster
    control planellm gatewayconsolepostgresqlclickhouse
    Your IdP - OIDC
    sso loginroles and bindings
    Your monitoring
    metricslogsyour inference endpoints

    OIDC login, your IdP

    Login runs through your OIDC provider. Roles, bindings, and API keys live in the identity services inside your cluster. No Cast AI account required.

    Governance that doesn't phone home

    Budgets, usage analytics, and the per-request audit trail are stored in the databases that ship with the stack. They keep working with no outbound connection.

    Your inference, fronted

    The gateway sits in front of what you already run - a vLLM server, an existing endpoint. Models you point it at are the models it serves.

    Start where it's easy. Move when the rules change.

    The API is the same in every pattern. Migration is a config change, not a rewrite.

    client-config.pyThe entire migration
    - base_url = "https://llm.kimchi.dev/openai/v1"
    + base_url = "https://llm.internal.yourcompany.com/v1"
    api_key = os.environ["KIMCHI_API_KEY"]
    model = "kimi-k3"
    One line. Same SDK, same budgets. Models depend on what you deploy.

    Cloud to hybrid

    Register your inference endpoints. Policy keeps deciding which requests stay in and which can still use serverless. Developers keep the same SDK. Only the endpoint may change.

    Hybrid to self-hosted

    Install the stack on your cluster, point clients at the internal endpoint. Budgets, analytics, and the audit trail come with it.

    Self-hosted to air-gapped

    Restrict inference to in-perimeter endpoints, isolate the network. That's the whole step.

    Bring governed AI coding inside your perimeter.

    We'll model the cost against your current API spend, walk through the patterns, and build a deployment plan around your requirements.