# Runpod

> Manage GPU Pods, Serverless endpoints, templates and network volumes with Runpod API v2, and submit, inspect, stream or cancel Serverless jobs.

App · 25 actions. Web page: https://docs.lodol.com/docs/integrations/runpod

## Actions

### Cancel Serverless job

Cancel Serverless job.

**Inputs**

- Endpoint (`endpoint_id`, text, required): Endpoint id
- Job ID (`job_id`, text, required): The job ID returned when the job was submitted.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): The token for the next page, or empty when complete.

**Example output**

```json
{
  "data": {
    "id": "example",
    "output": {},
    "status": "example",
    "stream": []
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Create a network volume

Create a network volume.

**Inputs**

- DataCenter (`dataCenter`, text, required): Data center in which to create the volume
- Name (`name`, text, required): Human-readable name
- Size (`size`, number, required): Storage to allocate in GB. From 10 to 4096.
- Type (`type`, text): Data center network volume storage type. One of: `STANDARD`, `HIGH_PERFORMANCE`.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "dataCenter": "example",
    "id": "example",
    "name": "example",
    "size": 10,
    "type": "STANDARD"
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Create a pod

Create a pod.

**Inputs**

- Container arguments (`container_args`, text): The container's command, as a single raw string. This is the field `entrypoint` and `cmd` encode into, exposed in its stored form. Two shapes are accepted. A bare shell string is treated as CMD and split into arguments, which is what the console's "Container start command" field writes. A JSON object of the form `{"entrypoint":[...],"cmd":[...]}` sets either or both explicitly. Responses always return both representations: `args` exactly as stored, plus the deconstructed `entrypoint` and `cmd`. Supplying `args` together with `entrypoint` or `cmd` is allowed only when they describe the same command, so a read-modify-write client can send back everything it received. Send `""` to clear, omit to leave unchanged.
- Cmd (`cmd`, list): Container CMD in exec form. When the image defines an ENTRYPOINT, this is the argument list passed to it. Encoded into the `args` field; supplying both is allowed only when they describe the same command. Send `[]` to clear, omit to leave unchanged.
- Entrypoint (`entrypoint`, list): Container ENTRYPOINT in exec form, overriding the image's own. Encoded into `args` field; supplying both is allowed only when they describe the same command. Send `[]` to clear, omit to leave unchanged.
- Disk (`disk`, number): Container disk in GB (ephemeral, wiped on restart). At least 1.
- Env (`env`, object): Environment variables as key-value pairs
- Image (`image`, text): Docker image reference
- Ports (`ports`, list): Exposed ports, formatted as port/protocol
- Registry (`registry`, text): Container registry credential ID (for private images)
- Name (`name`, text, required): Name
- Cloud (`cloud`, text): Cloud tier. - `SECURE` : Runpod-owned datacenter hardware - `COMMUNITY` : community-hosted hardware. One of: `SECURE`, `COMMUNITY`.
- CPU type (`cpu_id`, text): CPU flavor identifier, as returned by GET /v2/catalog/cpus.
- CPU vcpuCount (`cpu_vcpuCount`, number): Number of vCPUs. Must be valid for the selected CPU flavor and must be a power of two. At least 2.
- DataCenterIds (`dataCenterIds`, list): Preferred data centers for placement. Omit or pass an empty array to let the scheduler choose.
- GlobalNetworking (`globalNetworking`, true/false): Enable global networking, giving the pod a private IP reachable across data centers. Requires an NVIDIA GPU and a global-networking-enabled data center (both enforced upstream). See `GET /v2/catalog/datacenters` (`globalNetwork`) for eligible data centers.
- GPU type (`gpu_id`, text): GPU type identifier
- GPU count (`gpu_count`, number): Number of GPUs. At least 1.
- GPU allowedCudaVersions (`gpu_allowedCudaVersions`, list): Acceptable CUDA versions for the host machine, as `major.minor`. Omit to accept any version. Matching is exact, so a version no machine reports yields a capacity error rather than a fallback : discover valid values per GPU type via `GET /v2/catalog/gpus?include=AVAILABILITY&product=POD` (`cudaVersions`). A non-empty set is mutually exclusive with minCudaVersion (400 if both are sent). An explicit `[]` states no constraint, so it may accompany a floor.
- GPU minCudaVersion (`gpu_minCudaVersion`, text): Lowest acceptable CUDA version for the host machine, as `major.minor`, compared numerically rather than as a decimal : so 12.11 is above 12.2. Use this for an open-ended floor and allowedCudaVersions for an exact set. Mutually exclusive with a non-empty allowedCudaVersions (400 if both are sent); an explicit `[]` there states no constraint and may accompany this floor.
- GPU minRamPerGpu (`gpu_minRamPerGpu`, number): Minimum host system RAM in GB per requested GPU. This is a placement filter, not a resource request. The actual total allocated system RAM may be higher and is returned in gpu.memory. At least 1.
- GPU minVcpuCountPerGpu (`gpu_minVcpuCountPerGpu`, number): Minimum host vCPU count per requested GPU. This is a placement filter, not a resource request. The actual total allocated vCPU count may be higher and is returned in gpu.vcpuCount. At least 1.
- Mounts (`mounts`, object): Storage mounts attached to a pod. `persistent` is mutually exclusive with both `network` and `global` (enforced at the handler with 400 if combined). The `network` field is an array for forward compatibility with eventual multi-network-volume support, but `maxItems` is 1 today. PATCH semantics: - Omitting `mounts` or sending `{}` leaves the existing mount unchanged. - An explicit `network: []` is rejected with 400 (clearing mounts is not supported). - Mount kind is fixed at create : a PATCH that introduces a kind not present at create (persistent on a network pod, network on a persistent pod, or any mount on a previously-mountless pod) is rejected with 400. - The `volumeId` of a network mount is immutable; a PATCH that names a different `volumeId` is rejected with 400. - Partial mounts are not supported : every mount entry must include the full schema (`size` + `path` for persistent, `volumeId` + `path` for network). Missing required fields → 422.
- StartJupyter (`startJupyter`, true/false): Create-time flag telling the provisioner to start JupyterLab: injects a generated `JUPYTER_PASSWORD` environment variable, unless the request already sets one. Only images that honor the convention start Jupyter from it (Runpod official images do); expose `8888/http` in `ports` to reach it. Not part of the pod's readable config : never returned by GET and not changeable by PATCH.
- StartSsh (`startSsh`, true/false): Create-time flag telling the provisioner to set up SSH access: injects a `PUBLIC_KEY` environment variable carrying your account's registered SSH public keys, unless the request already sets one. **Requires registered keys** (`PUT /v2/account/ssh-keys`) : with none registered the flag does nothing and the pod has no SSH access. Only images that honor the convention start sshd from it (all Runpod official images do). Connect using the pod's `ssh` block; the `ssh.direct` variant additionally needs a `22/tcp` entry in `ports`. Not part of the pod's readable config : never returned by GET and not changeable by PATCH.
- TemplateId (`templateId`, text): ID of a pod template to base this pod on. The template is resolved at create time into the same container settings you could otherwise spread into this body (image, args, disk, ports, env, registry, persistent mount, startSsh, startJupyter, allowedCudaVersions); explicit body fields override the template's, except `env`, which is merged per key with body values winning. Sending either CUDA field (`gpu.allowedCudaVersions` or `gpu.minCudaVersion`) replaces the template's CUDA constraint entirely, and CPU pods ignore it (like the persistent mount). The template is a one-time source of settings: later template edits do not affect the pod, and the created pod does not retain a link to the template (`template` stays null). The template may be one of your own or a public catalog template : see `GET /v2/catalog/templates` (unknown or inaccessible ID → 404) : and must not be a serverless template (→ 422). CPU pods do not inherit a template's persistent mount.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "actions": [],
    "args": "example",
    "cloud": "SECURE",
    "cluster": {
      "id": "example",
      "ip": "example",
      "rank": null,
      "role": "SLURM_CONTROLLER"
    },
    "cmd": [],
    "cost": 0,
    "cpu": {
      "id": "example",
      "memory": 1,
      "vcpuCount": 2
    },
    "createdAt": "2026-01-01T00:00:00Z",
    "cudaVersion": null,
    "dataCenterId": null,
    "disk": 1,
    "entrypoint": [],
    "env": {},
    "globalNetworking": {
      "enabled": false,
      "internalDns": "example",
      "ip": "example"
    },
    "gpu": {
      "count": 1,
      "id": "example",
      "memory": 1,
      "vcpuCount": 1
    },
    "id": "example",
    "image": "example",
    "locked": false,
    "mounts": {
      "global": [],
      "network": [],
      "persistent": {
        "path": "example",
        "size": 10
      }
    },
    "name": "example",
    "ports": [],
    "registry": null,
    "runtime": null,
    "ssh": {
      "direct": null,
      "proxy": null
    },
    "startedAt": null,
    "status": "PROVISIONING",
    "template": null
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Create a serverless endpoint

Create a serverless endpoint.

**Inputs**

- Container arguments (`container_args`, text): The container's command, as a single raw string. This is the field `entrypoint` and `cmd` encode into, exposed in its stored form. Two shapes are accepted. A bare shell string is treated as CMD and split into arguments, which is what the console's "Container start command" field writes. A JSON object of the form `{"entrypoint":[...],"cmd":[...]}` sets either or both explicitly. Responses always return both representations: `args` exactly as stored, plus the deconstructed `entrypoint` and `cmd`. Supplying `args` together with `entrypoint` or `cmd` is allowed only when they describe the same command, so a read-modify-write client can send back everything it received. Send `""` to clear, omit to leave unchanged.
- Cmd (`cmd`, list): Container CMD in exec form. When the image defines an ENTRYPOINT, this is the argument list passed to it. Encoded into the `args` field; supplying both is allowed only when they describe the same command. Send `[]` to clear, omit to leave unchanged.
- Entrypoint (`entrypoint`, list): Container ENTRYPOINT in exec form, overriding the image's own. Encoded into `args` field; supplying both is allowed only when they describe the same command. Send `[]` to clear, omit to leave unchanged.
- Disk (`disk`, number): Container disk in GB (ephemeral, wiped on restart). At least 1.
- Env (`env`, object): Environment variables as key-value pairs
- Image (`image`, text): Docker image reference
- Ports (`ports`, list): Exposed ports, formatted as port/protocol
- Registry (`registry`, text): Container registry credential ID (for private images)
- Gpu (`gpu`, object): GPU request for an endpoint create. Carries the CUDA constraints, which live here rather than at the body's top level so they are unrepresentable on a CPU endpoint.
- Name (`name`, text, required): Name
- Scaling (`scaling`, object, required): Scale on queue wait time. Queue-based endpoints only.
- Type (`type`, text, required): Request-routing semantics for a modern serverless endpoint. - `QUEUE` : submit asynchronous or synchronous jobs through the managed queue. - `LOAD_BALANCER` : send requests directly to worker-defined HTTP paths. Configure via `env`: `PORT` (server port, default 80), `PORT_HEALTH` (health-check port, default 80), and `HEALTH_CHECK_PATH` (path the load balancer polls for worker health, default `/ping`). One of: `QUEUE`, `LOAD_BALANCER`.
- Cpu (`cpu`, list): Eligible CPU configurations for each worker. Memory is derived from the selected flavor's catalog RAM multiplier. Exact duplicate configurations are rejected; the same flavor may be listed at different vCPU counts.
- DataCenterIds (`dataCenterIds`, list): Preferred data centers for placement. Omit or pass an empty array to let the scheduler choose.
- Flashboot (`flashboot`, text): FlashBoot cold-start acceleration mode. - `OFF` : disabled - `FLASHBOOT` : enabled - `PRIORITY_FLASHBOOT` : enabled with priority capacity. One of: `OFF`, `FLASHBOOT`, `PRIORITY_FLASHBOOT`.
- NetworkVolumes (`networkVolumes`, list): Network volume IDs.
- Mounts (`mounts`, object): Volumes to mount on the endpoint's workers. A `network` or `global` list that is present replaces the endpoint's current list of that kind; an empty list detaches it, and an omitted list is left unchanged. Both kinds cannot be present on the endpoint at once, so to switch kinds send `[]` for one and the new list for the other in the same request. Changes apply to workers created afterwards. At most one global volume, GPU endpoints only. `network` may accompany the deprecated `networkVolumes` field only if the two agree exactly.
- TemplateId (`templateId`, text): ID of a serverless template to base this endpoint on. The template is resolved at create time into the same container settings you could otherwise spread into this body (image, args, disk, ports, env, registry); explicit body fields override the template's, except `env`, which is merged per key with body values winning. The template's allowedCudaVersions seeds `gpu.allowedCudaVersions` when the body omits it : but only for a GPU create, since a CPU endpoint has no gpu block to seed into, and not when the body sets `gpu.minCudaVersion`, since seeding a set beside a floor would manufacture the mutual-exclusion 400 from a valid request. Its pod-specific startSsh/startJupyter flags are ignored. Later template edits do not affect the endpoint. The template may be one of your own or a public catalog template : see `GET /v2/catalog/templates` (unknown or inaccessible ID → 404) : and must be a serverless template (→ 422).
- Timeout (`timeout`, number): Timeout
- Workers (`workers`, object): Workers

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "args": "example",
    "cmd": [],
    "cpu": [],
    "createdAt": "2026-01-01T00:00:00Z",
    "dataCenterIds": [],
    "disk": 1,
    "entrypoint": [],
    "env": {},
    "flashboot": "OFF",
    "gpu": null,
    "id": "example",
    "image": "example",
    "mounts": {
      "global": [],
      "network": []
    },
    "name": "example",
    "networkVolumes": [],
    "ports": [],
    "registry": null,
    "requestUrls": {
      "cancel": "example",
      "health": "example",
      "purgeQueue": "example",
      "retry": "example",
      "run": "example",
      "runSync": "example",
      "status": "example",
      "stream": "example"
    },
    "scaling": {
      "queueDelay": 0.5,
      "type": "QUEUE_DELAY"
    },
    "timeout": 0,
    "type": "QUEUE",
    "workers": {
      "idleTimeout": 1,
      "max": 0,
      "min": 0
    }
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Create a template

Create a template.

**Inputs**

- Container arguments (`container_args`, text): The container's command, as a single raw string. This is the field `entrypoint` and `cmd` encode into, exposed in its stored form. Two shapes are accepted. A bare shell string is treated as CMD and split into arguments, which is what the console's "Container start command" field writes. A JSON object of the form `{"entrypoint":[...],"cmd":[...]}` sets either or both explicitly. Responses always return both representations: `args` exactly as stored, plus the deconstructed `entrypoint` and `cmd`. Supplying `args` together with `entrypoint` or `cmd` is allowed only when they describe the same command, so a read-modify-write client can send back everything it received. Send `""` to clear, omit to leave unchanged.
- Cmd (`cmd`, list): Container CMD in exec form. When the image defines an ENTRYPOINT, this is the argument list passed to it. Encoded into the `args` field; supplying both is allowed only when they describe the same command. Send `[]` to clear, omit to leave unchanged.
- Entrypoint (`entrypoint`, list): Container ENTRYPOINT in exec form, overriding the image's own. Encoded into `args` field; supplying both is allowed only when they describe the same command. Send `[]` to clear, omit to leave unchanged.
- Disk (`disk`, number): Container disk in GB (ephemeral, wiped on restart). At least 1.
- Env (`env`, object): Environment variables as key-value pairs
- Image (`image`, text, required): Docker image reference
- Ports (`ports`, list): Exposed ports, formatted as port/protocol
- Registry (`registry`, text): Container registry credential ID (for private images)
- Name (`name`, text, required): Name
- AllowedCudaVersions (`allowedCudaVersions`, list): Acceptable CUDA versions for containers created from this template, as `major.minor`. Omit to accept any version : see the same field on `createPod` for matching semantics. Expanded into GPU pod and serverless endpoint creates; CPU pods ignore it.
- Category (`category`, text): Controls how the template is grouped and filtered in the Runpod console. It does not affect hardware selection, scheduling, or billing. - `CPU` : CPU-only workloads - `NVIDIA` : NVIDIA GPU workloads - `AMD` : AMD GPU workloads. One of: `CPU`, `NVIDIA`, `AMD`.
- Mounts (`mounts`, object): Storage mounts attached to a template. Templates support only a single persistent mount today; any `network` property is rejected with 422 by the schema validator. PATCH semantics: omitting `mounts` or sending `{}` leaves the existing mount unchanged.
- Public (`public`, true/false): Public
- Serverless (`serverless`, true/false): Serverless
- StartJupyter (`startJupyter`, true/false): Start JupyterLab in containers created from this template: injects a generated `JUPYTER_PASSWORD` environment variable, unless `env` already sets one. Only images that honor the convention start Jupyter from it (Runpod official images do); expose `8888/http` in `ports` to reach it. Defaults to `true` when omitted, matching console-created templates.
- StartSsh (`startSsh`, true/false): Provision SSH access in containers created from this template: injects a `PUBLIC_KEY` environment variable carrying the deployer's registered SSH public keys (`PUT /v2/account/ssh-keys` : with none registered the flag does nothing), unless `env` already sets one. Only images that honor the convention start sshd from it (all Runpod official images do); direct SSH also needs a `22/tcp` entry in `ports`. Defaults to `true` when omitted, matching console-created templates.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "allowedCudaVersions": [],
    "args": "example",
    "category": "CPU",
    "cmd": [],
    "disk": 1,
    "entrypoint": [],
    "env": {},
    "id": "example",
    "image": "example",
    "mounts": {
      "persistent": {
        "path": "example",
        "size": 10
      }
    },
    "name": "example",
    "ports": [],
    "public": false,
    "registry": null,
    "serverless": false,
    "startJupyter": false,
    "startSsh": false
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Delete a network volume

Delete a network volume.

**Inputs**

- Network volume (`id`, text, required): Network volume identifier

**Outputs**

- `data` (null): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": null,
  "has_more": false,
  "next_cursor": "example"
}
```

### Delete a serverless endpoint

Delete a serverless endpoint.

**Inputs**

- Endpoint (`id`, text, required): Serverless endpoint identifier

**Outputs**

- `data` (null): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": null,
  "has_more": false,
  "next_cursor": "example"
}
```

### Delete a template

Delete a template.

**Inputs**

- Template (`id`, text, required): Id

**Outputs**

- `data` (null): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": null,
  "has_more": false,
  "next_cursor": "example"
}
```

### Get a network volume

Get a network volume.

**Inputs**

- Network volume (`id`, text, required): Network volume identifier

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "dataCenter": "example",
    "id": "example",
    "name": "example",
    "size": 10,
    "type": "STANDARD"
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Get a pod

Get a pod.

**Inputs**

- Pod (`id`, text, required): Pod identifier

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "actions": [],
    "args": "example",
    "cloud": "SECURE",
    "cluster": {
      "id": "example",
      "ip": "example",
      "rank": null,
      "role": "SLURM_CONTROLLER"
    },
    "cmd": [],
    "cost": 0,
    "cpu": {
      "id": "example",
      "memory": 1,
      "vcpuCount": 2
    },
    "createdAt": "2026-01-01T00:00:00Z",
    "cudaVersion": null,
    "dataCenterId": null,
    "disk": 1,
    "entrypoint": [],
    "env": {},
    "globalNetworking": {
      "enabled": false,
      "internalDns": "example",
      "ip": "example"
    },
    "gpu": {
      "count": 1,
      "id": "example",
      "memory": 1,
      "vcpuCount": 1
    },
    "id": "example",
    "image": "example",
    "locked": false,
    "mounts": {
      "global": [],
      "network": [],
      "persistent": {
        "path": "example",
        "size": 10
      }
    },
    "name": "example",
    "ports": [],
    "registry": null,
    "runtime": null,
    "ssh": {
      "direct": null,
      "proxy": null
    },
    "startedAt": null,
    "status": "PROVISIONING",
    "template": null
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Get a serverless endpoint

Get a serverless endpoint.

**Inputs**

- Endpoint (`id`, text, required): Serverless endpoint identifier

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "args": "example",
    "cmd": [],
    "cpu": [],
    "createdAt": "2026-01-01T00:00:00Z",
    "dataCenterIds": [],
    "disk": 1,
    "entrypoint": [],
    "env": {},
    "flashboot": "OFF",
    "gpu": null,
    "id": "example",
    "image": "example",
    "mounts": {
      "global": [],
      "network": []
    },
    "name": "example",
    "networkVolumes": [],
    "ports": [],
    "registry": null,
    "requestUrls": {
      "cancel": "example",
      "health": "example",
      "purgeQueue": "example",
      "retry": "example",
      "run": "example",
      "runSync": "example",
      "status": "example",
      "stream": "example"
    },
    "scaling": {
      "queueDelay": 0.5,
      "type": "QUEUE_DELAY"
    },
    "timeout": 0,
    "type": "QUEUE",
    "workers": {
      "idleTimeout": 1,
      "max": 0,
      "min": 0
    }
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Get a template

Get a template.

**Inputs**

- Template (`id`, text, required): Id

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "allowedCudaVersions": [],
    "args": "example",
    "category": "CPU",
    "cmd": [],
    "disk": 1,
    "entrypoint": [],
    "env": {},
    "id": "example",
    "image": "example",
    "mounts": {
      "persistent": {
        "path": "example",
        "size": 10
      }
    },
    "name": "example",
    "ports": [],
    "public": false,
    "registry": null,
    "serverless": false,
    "startJupyter": false,
    "startSsh": false
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Get endpoint health

Get endpoint health.

**Inputs**

- Endpoint (`endpoint_id`, text, required): Endpoint id

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): The token for the next page, or empty when complete.

**Example output**

```json
{
  "data": {},
  "has_more": false,
  "next_cursor": "example"
}
```

### Get Serverless job

Get Serverless job.

**Inputs**

- Endpoint (`endpoint_id`, text, required): Endpoint id
- Job ID (`job_id`, text, required): The job ID returned when the job was submitted.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): The token for the next page, or empty when complete.

**Example output**

```json
{
  "data": {
    "id": "example",
    "output": {},
    "status": "example",
    "stream": []
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Get streamed job results

Get streamed job results.

**Inputs**

- Endpoint (`endpoint_id`, text, required): Endpoint id
- Job ID (`job_id`, text, required): The job ID returned when the job was submitted.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): The token for the next page, or empty when complete.

**Example output**

```json
{
  "data": {
    "id": "example",
    "output": {},
    "status": "example",
    "stream": []
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### List network volumes

List network volumes.

**Inputs**

- Cursor (`cursor`, text): Opaque resume cursor : pass the previous response's `pagination.nextCursor` through verbatim; omit for the first page. A cursor is only valid for the operation and parameters that issued it; a malformed or foreign cursor is rejected with 422.
- Limit (`limit`, number): Page size, 1to1000. Defaults to 1000 when omitted. From 1 to 1000.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "networkVolumes": [],
    "pagination": {
      "hasNextPage": false,
      "nextCursor": null
    }
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### List pods

List pods.

**Inputs**

- IncludeClusterPods (`includeClusterPods`, true/false): Include cluster member pods in the result. Defaults to false.
- Cursor (`cursor`, text): Opaque resume cursor : pass the previous response's `pagination.nextCursor` through verbatim; omit for the first page. A cursor is only valid for the operation and parameters that issued it; a malformed or foreign cursor is rejected with 422.
- Limit (`limit`, number): Page size, 1to1000. Defaults to 1000 when omitted. From 1 to 1000.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "pagination": {
      "hasNextPage": false,
      "nextCursor": null
    },
    "pods": []
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### List serverless endpoint workers

List serverless endpoint workers.

**Inputs**

- Endpoint (`id`, text, required): Serverless endpoint identifier

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "endpointVersion": null,
    "summary": {
      "idle": 0,
      "initializing": 0,
      "running": 0,
      "throttled": 0,
      "total": 0,
      "unhealthy": 0
    },
    "workers": []
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### List serverless endpoints

List serverless endpoints.

**Inputs**

- Cursor (`cursor`, text): Opaque resume cursor : pass the previous response's `pagination.nextCursor` through verbatim; omit for the first page. A cursor is only valid for the operation and parameters that issued it; a malformed or foreign cursor is rejected with 422.
- Limit (`limit`, number): Page size, 1to1000. Defaults to 1000 when omitted. From 1 to 1000.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "endpoints": [],
    "pagination": {
      "hasNextPage": false,
      "nextCursor": null
    }
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### List templates

List templates.

**Inputs**

- Cursor (`cursor`, text): Opaque resume cursor : pass the previous response's `pagination.nextCursor` through verbatim; omit for the first page. A cursor is only valid for the operation and parameters that issued it; a malformed or foreign cursor is rejected with 422.
- Limit (`limit`, number): Page size, 1to1000. Defaults to 1000 when omitted. From 1 to 1000.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "pagination": {
      "hasNextPage": false,
      "nextCursor": null
    },
    "templates": []
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Run Serverless job

Run Serverless job.

**Inputs**

- Endpoint (`endpoint_id`, text, required): Endpoint id
- Job input (`job_input`, object, required): The input object expected by your endpoint’s worker.
- Webhook (`webhook`, text): HTTPS address to receive the completed job.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): The token for the next page, or empty when complete.

**Example output**

```json
{
  "data": {
    "id": "example",
    "output": {},
    "status": "example",
    "stream": []
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Run Serverless job and wait

Submit a Serverless job and wait briefly for a result. If it is still running, use the returned job ID with Get Serverless job; resubmitting creates another paid job.

**Inputs**

- Endpoint (`endpoint_id`, text, required): Endpoint id
- Job input (`job_input`, object, required): The input object expected by your endpoint’s worker.
- Wait (`wait`, number): Maximum wait in milliseconds. A job may continue in the background after this wait. From 1000 to 50000.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): The token for the next page, or empty when complete.

**Example output**

```json
{
  "data": {
    "id": "example",
    "output": {},
    "status": "example",
    "stream": []
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Terminate a pod

Terminate a pod.

**Inputs**

- Pod (`id`, text, required): Pod identifier

**Outputs**

- `data` (null): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": null,
  "has_more": false,
  "next_cursor": "example"
}
```

### Trigger a pod state transition

Trigger a pod state transition.

**Inputs**

- Pod (`id`, text, required): Id
- Action (`pod_action`, text, required): State transition to trigger on a pod. One of: `start`, `stop`, `restart`, `terminate`.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "actions": [],
    "args": "example",
    "cloud": "SECURE",
    "cluster": {
      "id": "example",
      "ip": "example",
      "rank": null,
      "role": "SLURM_CONTROLLER"
    },
    "cmd": [],
    "cost": 0,
    "cpu": {
      "id": "example",
      "memory": 1,
      "vcpuCount": 2
    },
    "createdAt": "2026-01-01T00:00:00Z",
    "cudaVersion": null,
    "dataCenterId": null,
    "disk": 1,
    "entrypoint": [],
    "env": {},
    "globalNetworking": {
      "enabled": false,
      "internalDns": "example",
      "ip": "example"
    },
    "gpu": {
      "count": 1,
      "id": "example",
      "memory": 1,
      "vcpuCount": 1
    },
    "id": "example",
    "image": "example",
    "locked": false,
    "mounts": {
      "global": [],
      "network": [],
      "persistent": {
        "path": "example",
        "size": 10
      }
    },
    "name": "example",
    "ports": [],
    "registry": null,
    "runtime": null,
    "ssh": {
      "direct": null,
      "proxy": null
    },
    "startedAt": null,
    "status": "PROVISIONING",
    "template": null
  },
  "has_more": false,
  "next_cursor": "example"
}
```

### Update a template

Update a template.

**Inputs**

- Template (`id`, text, required): Id
- Container arguments (`container_args`, text): The container's command, as a single raw string. This is the field `entrypoint` and `cmd` encode into, exposed in its stored form. Two shapes are accepted. A bare shell string is treated as CMD and split into arguments, which is what the console's "Container start command" field writes. A JSON object of the form `{"entrypoint":[...],"cmd":[...]}` sets either or both explicitly. Responses always return both representations: `args` exactly as stored, plus the deconstructed `entrypoint` and `cmd`. Supplying `args` together with `entrypoint` or `cmd` is allowed only when they describe the same command, so a read-modify-write client can send back everything it received. Send `""` to clear, omit to leave unchanged.
- Cmd (`cmd`, list): Container CMD in exec form. When the image defines an ENTRYPOINT, this is the argument list passed to it. Encoded into the `args` field; supplying both is allowed only when they describe the same command. Send `[]` to clear, omit to leave unchanged.
- Entrypoint (`entrypoint`, list): Container ENTRYPOINT in exec form, overriding the image's own. Encoded into `args` field; supplying both is allowed only when they describe the same command. Send `[]` to clear, omit to leave unchanged.
- Disk (`disk`, number): Container disk in GB (ephemeral, wiped on restart). At least 1.
- Env (`env`, object): Environment variables as key-value pairs
- Image (`image`, text): Docker image reference
- Ports (`ports`, list): Exposed ports, formatted as port/protocol
- Registry (`registry`, text): Container registry credential ID (for private images)
- AllowedCudaVersions (`allowedCudaVersions`, list): Acceptable CUDA versions for pods created from this template. An explicit `[]` clears the constraint; omitting the field leaves it unchanged.
- Category (`category`, text): Controls how the template is grouped and filtered in the Runpod console. It does not affect hardware selection, scheduling, or billing. - `CPU` : CPU-only workloads - `NVIDIA` : NVIDIA GPU workloads - `AMD` : AMD GPU workloads. One of: `CPU`, `NVIDIA`, `AMD`.
- Mounts (`mounts`, object): Storage mounts attached to a template. Templates support only a single persistent mount today; any `network` property is rejected with 422 by the schema validator. PATCH semantics: omitting `mounts` or sending `{}` leaves the existing mount unchanged.
- Name (`name`, text): Name
- Public (`public`, true/false): Public
- Serverless (`serverless`, true/false): Serverless
- StartJupyter (`startJupyter`, true/false): Start JupyterLab at container startup (`JUPYTER_PASSWORD` env injection). See the create-time field for details.
- StartSsh (`startSsh`, true/false): Provision SSH access at container startup (`PUBLIC_KEY` env injection). See the create-time field for details.

**Outputs**

- `data` (object): The response returned by the service.
- `has_more` (true/false): Whether another page is available.
- `next_cursor` (text): Continuation token for the next page.

**Example output**

```json
{
  "data": {
    "allowedCudaVersions": [],
    "args": "example",
    "category": "CPU",
    "cmd": [],
    "disk": 1,
    "entrypoint": [],
    "env": {},
    "id": "example",
    "image": "example",
    "mounts": {
      "persistent": {
        "path": "example",
        "size": 10
      }
    },
    "name": "example",
    "ports": [],
    "public": false,
    "registry": null,
    "serverless": false,
    "startJupyter": false,
    "startSsh": false
  },
  "has_more": false,
  "next_cursor": "example"
}
```
