# Run platform URL: https://docs.hanzo.ai/docs/openapi/platform/post-platform-run > Runs a container image and gives back a URL. `POST /v1/platform/run` > [Platform →](https://docs.hanzo.ai/docs/openapi/platform) · [All API references →](https://docs.hanzo.ai/docs/openapi) · [Get started →](https://docs.hanzo.ai/docs/guides) | | | | ------------- | -------------------------------------- | | **Address** | `https://api.hanzo.ai/v1/platform/run` | | **Method** | `POST` | | **Operation** | `post_platform_run` | | **Auth** | `Authorization: Bearer $HANZO_API_KEY` | Runs a container image and gives back a URL. The one-call shortcut over project → app → deploy: give it a `name` and an `image` and it creates or updates an image-source application in your org's DEFAULT project, deploys it through the same operator Service-CR writer everything else uses, and answers its id, name, live URL, status and shape. Re-running the same name UPDATES it in place, so the call is idempotent by name. What it produces is a first-class application, not a special object: it is listable, stoppable and redeployable through the /v1/platform routes like any other app. `minScale` is the replica floor. `maxScale` above it declares an autoscaling ceiling; `maxScale: 0` means no autoscaler at all — a fixed run at the floor. Both are clamped to the deployment's limits. `runtime` and `shape` are accepted for the client contract and echoed back: the image is the runtime unit and sizing is the operator's default. It is BILLING-GATED before it touches the cluster: a flat per-run fee is authorized against the org's own prepaid balance first, so an org that cannot pay is refused without anything being created. An unreachable cluster is 503 — a run never reports a URL it did not create. Secret env is sealed into KMS and fails closed without it. Requires a validated principal; 401 without one. The org is resolved from that validated identity and is what both pays and owns the namespace — it is never read from the body. ## Request 12 fields, body `application/json` (required). | Field | In | Type | Required | Description | | -------------- | ---- | ---------------------- | -------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | `env` | body | platform.EnvVarJSON\[] | — | Env is the run's environment. | | `env[].key` | body | string | — | Key is the variable's name in the container, which must match `^[A-Za-z_][A-Za-z0-9_]*$`. | | `env[].secret` | body | boolean | — | Secret says the value lives in KMS and never in the database. | | `env[].value` | body | string | — | Value is the plaintext, and it is WRITE-ONLY once the entry is secret: a sealed value reads back as "", and sending "" again KEEPS what is sealed rather than wiping it. | | `gpu` | body | integer (int64) | — | GPU is how many GPUs the run asks for; a negative value is 400. | | `image` | body | string | — | Image is the container image to run. | | `maxScale` | body | integer (int64) | — | MaxScale above the floor declares an autoscaling ceiling; 0 means no autoscaler at all — a fixed run at the floor. | | `minScale` | body | integer (int64) | — | MinScale is the replica floor, clamped to the deployment's limit. | | `name` | body | string | — | Name is the run's name, and the slug is derived from it. | | `port` | body | integer (int64) | — | Port is the container port the run listens on. | | `runtime` | body | string | — | Runtime is accepted for the client contract and echoed nowhere: the image IS the runtime unit. | | `shape` | body | string | — | Shape is a compute size label, echoed back; sizing is the operator's default. | ## Response | Status | Body | Meaning | | --------- | ---------------- | -------- | | `202` | platform.runView | accepted | | `default` | problem-details | refused | `202` body — 5 fields. | Field | In | Type | Always | Description | | -------- | ---- | ------ | ------ | ------------------------------------------------------------------ | | `id` | body | string | — | ID is the application id the run created or converged. | | `name` | body | string | — | Name is the run's name, as stored. | | `shape` | body | string | — | Shape is the compute size label the request asked for, or "auto". | | `status` | body | string | — | Status is the application's state — `deploying` on a fresh accept. | | `url` | body | string | — | URL is the run's live HTTPS address. | Failure carries the platform error shape — see [Errors](https://docs.hanzo.ai/docs/errors). ## Examples **CLI** ```bash hanzo platform run ``` **SDK** **.ts** ```ts import { Configuration, PlatformApi } from 'hanzoai'; const api = new PlatformApi(new Configuration({ accessToken: process.env.HANZO_API_KEY })); const { data } = await api.postPlatformRun({ env: [{"key":"","secret":false,"value":""}], gpu: 0 }); ``` **.py** ```python from hanzoai.cloud import ApiClient, Configuration from hanzoai.cloud.api import PlatformApi client = ApiClient(Configuration(access_token=os.environ["HANZO_API_KEY"])) result = PlatformApi(client).post_platform_run(env=[{"key":"","secret":false,"value":""}], gpu=0) ``` **.go** ```go cfg := hanzoai.NewConfiguration() cfg.AddDefaultHeader("Authorization", "Bearer "+os.Getenv("HANZO_API_KEY")) client := hanzoai.NewAPIClient(cfg) resp, _, err := client.PlatformAPI.PostPlatformRun(context.Background()).Execute() if err != nil { return err } ``` **.rs** ```rust use hanzo_client::apis::{configuration::Configuration, platform_api}; let mut cfg = Configuration::new(); cfg.bearer_access_token = std::env::var("HANZO_API_KEY").ok(); let result = platform_api::post_platform_run(&cfg, Default::default()).await?; ``` **.java** ```java import ai.hanzo.cloud.ApiClient; import ai.hanzo.cloud.api.PlatformApi; ApiClient client = new ApiClient(); client.setBearerToken(System.getenv("HANZO_API_KEY")); var result = new PlatformApi(client).postPlatformRun(); ``` **HTTP** ```bash curl -X POST https://api.hanzo.ai/v1/platform/run \ -H "Authorization: Bearer $HANZO_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "env": [ { "key": "", "secret": false, "value": "" } ], "gpu": 0 }' ``` **MCP** MCP reaches **platform** through the `platform` tool, which names its 38 operations with its own verbs — this one among them, under a name only MCP declares. `describe` explains any of them: ```bash curl -X POST https://api.hanzo.ai/v1/mcp \ -H "Content-Type: application/json" \ -d '{ "jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "describe", "arguments": { "op": "get_build" } } }' ``` --- [Platform API](https://docs.hanzo.ai/docs/openapi/platform) · [All Hanzo APIs](https://docs.hanzo.ai/docs/openapi) · [Interactive reference](https://docs.hanzo.ai/docs/openapi/reference)