CoreWeave
App · 20 actions
Manage inference deployments, gateways, capacity claims and checkpoint hot-loads.
Actions
| Action | What it does |
|---|---|
| Cancel checkpoint hot-load | Cancels a hot-load. |
| Create capacity claim | Creates a new CapacityClaim that reserves GPU hardware for inference deployments. |
| Create checkpoint hot-load | Triggers a checkpoint hot-load on a deployment. |
| Create deployment | Creates a new model deployment associated with one or more gateways. |
| Create gateway | Creates a new inference gateway. |
| Delete capacity claim | Deletes a CapacityClaim by ID. |
| Delete deployment | Deletes a deployment by ID. |
| Delete gateway | Deletes a gateway by ID. |
| Get capacity claim | Retrieves a single CapacityClaim by its UUID. |
| Get checkpoint hot-load | Returns a hot-load by ID, including its resolved snapshot chain and aggregate state. |
| Get deployment | Retrieves a single deployment by ID. |
| Get deployment parameters | Returns the available gateways, runtime engines and versions, runtime config keys, and instance types for use when creating a deployment. |
| Get gateway | Retrieves a single gateway by its UUID. |
| List capacity claims | Returns every CapacityClaim that the caller can read. |
| List checkpoint hot-loads | Returns the caller's hot-loads, newest first, optionally filtered to a single deployment. |
| List deployments | Returns every deployment that the caller can read, optionally filtered by parent gateway or update timestamp. |
| List gateways | Returns every inference gateway that the caller can read, optionally filtered by update timestamp. |
| List hot-load replicas | Returns the replicas of one hot-load, ordered by zone and name, in pages. |
| Update deployment | Updates an existing deployment. |
| Update gateway | Updates an existing gateway. |
Cancel checkpoint hot-load
Cancels a hot-load. This operation is not yet implemented and currently returns an Unimplemented error.
Inputs
- ID
idtextRequired - The ID of the hot-load to cancel
Outputs
dataobject- Response for CancelHotLoad
Example output
{
"data": {
"hotLoad": {
"spec": {
"deploymentId": "example",
"id": "example",
"identity": "example",
"incrementalSnapshotMetadata": {},
"organizationId": "example",
"promptCachePolicy": "example",
"snapshotChain": [],
"type": "example"
},
"status": {
"canceledAt": "example",
"conditions": [],
"createdAt": "example",
"finishedAt": "example",
"state": "example",
"updatedAt": "example"
}
}
}
}Create capacity claim
Creates a new CapacityClaim that reserves GPU hardware for inference deployments.
Inputs
- ID
idtext - The unique identifier of the CapacityClaim, UUID format
- Name
nametext - The human readable name of the CapacityClaim
- Resources
resourcesobjectRequired - The specification for the resources to reserve
Outputs
dataobject- Response for CreateCapacityClaim
Example output
{
"data": {
"capacityClaim": {
"spec": {
"id": "example",
"name": "example",
"organizationId": "example",
"resources": {}
},
"status": {
"allocatedInstances": 1,
"conditions": [],
"createdAt": "example",
"pendingInstances": 1,
"status": "example",
"updatedAt": "example"
}
}
}
}Create checkpoint hot-load
Triggers a checkpoint hot-load on a deployment. The operation runs asynchronously, so a successful response means the request was accepted, not that the weights are loaded.
Inputs
- Deployment
deployment_idtextRequired - The ID of the deployment to hot-load
- Hot load
hot_load_idtext - The ID to use for the hot-load, which becomes its permanent identifier. Must be a UUID. If omitted, one is generated. Creating a second hot-load with an ID that already exists in the caller's organization fails with ALREADY_EXISTS; an ID that is unavailable for any other…
- Identity
identitytextRequired - The identity of the target snapshot within the deployment. Must be a single path segment. Multiple hot-load operations may target the same snapshot identity.
- Type
typetextRequired - The type of the target snapshot. SNAPSHOT_TYPE_FULL carries a complete set of model weights and starts a new snapshot chain. SNAPSHOT_TYPE_INCREMENTAL is a delta against an earlier snapshot and requires incrementalSnapshotMetadata. This field is required; the zero value is…One of:
SNAPSHOT_TYPE_FULL(Snapshot type full),SNAPSHOT_TYPE_INCREMENTAL(Snapshot type incremental). - Incremental snapshot metadata
incremental_snapshot_metadataobject - Delta metadata, required when type is SNAPSHOT_TYPE_INCREMENTAL
- Prompt cache policy
prompt_cache_policytext - How the prompt cache is handled once the swap completes. PROMPT_CACHE_POLICY_RESET_ALL prevents reuse of any entry computed before the swap. PROMPT_CACHE_POLICY_PRESERVE allows reuse, knowingly stale under the new weights. The policy applies to requests admitted after the…One of:
PROMPT_CACHE_POLICY_UNSPECIFIED(Prompt cache policy unspecified),PROMPT_CACHE_POLICY_RESET_ALL(Prompt cache policy reset all),PROMPT_CACHE_POLICY_PRESERVE(Prompt cache policy preserve).
Outputs
dataobject- Response for CreateHotLoad
Example output
{
"data": {
"hotLoad": {
"spec": {
"deploymentId": "example",
"id": "example",
"identity": "example",
"incrementalSnapshotMetadata": {},
"organizationId": "example",
"promptCachePolicy": "example",
"snapshotChain": [],
"type": "example"
},
"status": {
"canceledAt": "example",
"conditions": [],
"createdAt": "example",
"finishedAt": "example",
"state": "example",
"updatedAt": "example"
}
}
}
}Create deployment
Creates a new model deployment associated with one or more gateways.
Inputs
- ID
idtext - The unique identifier of the deployment, UUID format
- Name
nametextRequired - The name of the deployment
- Gateway IDS
gateway_idslistRequired - The gateways to associate the deployment with
- Runtime
runtimeobjectRequired - Runtime selection and configuration
- Resources
resourcesobjectRequired - Resource configuration for the deployment
- Model
modelobjectRequired - The model configuration
- Autoscaling
autoscalingobjectRequired - The autoscaling configuration
- Traffic
trafficobjectRequired - The traffic configuration for the deployment
- Disabled
disabledtrue/false - Disable the deployment
- Hot load
hot_loadobject - The hot-load configuration for the deployment. Set it to enable checkpoint hot-loading; it is immutable after creation.
- Pd
pdobject - Prefill/Decode disaggregation configuration. Top-level resources and autoscaling configure the decode pool; this field configures the prefill pool. Only supported by dynamo-vllm when the pdDisaggEnabled feature flag is enabled.
Outputs
dataobject- Response for CreateDeployment
Example output
{
"data": {
"deployment": {
"spec": {
"autoscaling": {},
"disabled": false,
"gatewayIds": [],
"hotLoad": {},
"id": "example",
"model": {},
"name": "example",
"organizationId": "example",
"pd": {},
"resources": {},
"runtime": {},
"traffic": {}
},
"status": {
"conditions": [],
"createdAt": "example",
"specRevision": "example",
"status": "example",
"updatedAt": "example"
}
}
}
}Create gateway
Creates a new inference gateway. Each gateway must be configured with exactly one authentication type and exactly one routing strategy.
Inputs
- ID
idtext - The unique identifier of the gateway, UUID format
- Name
nametextRequired - The human readable name of the gateway. Capped at 38 characters so the derived production FQDN (..gw.cwinference.com) fits within the 64-character X.509 Common Name limit.
- Zones
zoneslist - The zones to make the gateway available in, limits where deployments associated with the gateway may exist, no zones means all may be used.
- Core weave auth
core_weave_authobject - CoreWeave IAM
- Weights and biases auth
weights_and_biases_authobject - Weights & Biases
- Endpoint configuration
endpoint_configurationobject - Additional endpoint configuration options
- Body based routing
body_based_routingobject - Body based routing
- Header based routing
header_based_routingobject - Header based routing
- Path based routing
path_based_routingobject - Path based routing
Outputs
dataobject- Response for CreateGateway
Example output
{
"data": {
"gateway": {
"spec": {
"bodyBasedRouting": {},
"coreWeaveAuth": {},
"endpointConfiguration": {},
"headerBasedRouting": {},
"id": "example",
"name": "example",
"organizationId": "example",
"pathBasedRouting": {},
"weightsAndBiasesAuth": {},
"zones": []
},
"status": {
"conditions": [],
"createdAt": "example",
"endpoints": [],
"status": "example",
"updatedAt": "example"
}
}
}
}Delete capacity claim
Deletes a CapacityClaim by ID. Returns the deleted CapacityClaim.
Inputs
- ID
idtextRequired - The unique identifier of the CapacityClaim to delete, UUID format
Outputs
dataobject- Response for DeleteCapacityClaim
Example output
{
"data": {
"capacityClaim": {
"spec": {
"id": "example",
"name": "example",
"organizationId": "example",
"resources": {}
},
"status": {
"allocatedInstances": 1,
"conditions": [],
"createdAt": "example",
"pendingInstances": 1,
"status": "example",
"updatedAt": "example"
}
}
}
}Delete deployment
Deletes a deployment by ID. Returns the deleted deployment.
Inputs
- ID
idtextRequired - The ID of the deployment to delete
Outputs
dataobject- Response for DeleteDeployment
Example output
{
"data": {
"deployment": {
"spec": {
"autoscaling": {},
"disabled": false,
"gatewayIds": [],
"hotLoad": {},
"id": "example",
"model": {},
"name": "example",
"organizationId": "example",
"pd": {},
"resources": {},
"runtime": {},
"traffic": {}
},
"status": {
"conditions": [],
"createdAt": "example",
"specRevision": "example",
"status": "example",
"updatedAt": "example"
}
}
}
}Delete gateway
Deletes a gateway by ID. Returns the deleted gateway.
Inputs
- ID
idtextRequired - The unique identifier of the gateway to delete, UUID format
Outputs
dataobject- Response for DeleteGateway
Example output
{
"data": {
"gateway": {
"spec": {
"bodyBasedRouting": {},
"coreWeaveAuth": {},
"endpointConfiguration": {},
"headerBasedRouting": {},
"id": "example",
"name": "example",
"organizationId": "example",
"pathBasedRouting": {},
"weightsAndBiasesAuth": {},
"zones": []
},
"status": {
"conditions": [],
"createdAt": "example",
"endpoints": [],
"status": "example",
"updatedAt": "example"
}
}
}
}Get capacity claim
Retrieves a single CapacityClaim by its UUID.
Inputs
- ID
idtextRequired - The unique identifier of the CapacityClaim to get, UUID format
Outputs
dataobject- Response for GetCapacityClaim
Example output
{
"data": {
"capacityClaim": {
"spec": {
"id": "example",
"name": "example",
"organizationId": "example",
"resources": {}
},
"status": {
"allocatedInstances": 1,
"conditions": [],
"createdAt": "example",
"pendingInstances": 1,
"status": "example",
"updatedAt": "example"
}
}
}
}Get checkpoint hot-load
Returns a hot-load by ID, including its resolved snapshot chain and aggregate state.
Inputs
- ID
idtextRequired - The ID of the hot-load
Outputs
dataobject- Response for GetHotLoad
Example output
{
"data": {
"hotLoad": {
"spec": {
"deploymentId": "example",
"id": "example",
"identity": "example",
"incrementalSnapshotMetadata": {},
"organizationId": "example",
"promptCachePolicy": "example",
"snapshotChain": [],
"type": "example"
},
"status": {
"canceledAt": "example",
"conditions": [],
"createdAt": "example",
"finishedAt": "example",
"state": "example",
"updatedAt": "example"
}
}
}
}Get deployment
Retrieves a single deployment by ID.
Inputs
- ID
idtextRequired - The ID of the deployment to get
Outputs
dataobject- Response for GetDeployment
Example output
{
"data": {
"deployment": {
"spec": {
"autoscaling": {},
"disabled": false,
"gatewayIds": [],
"hotLoad": {},
"id": "example",
"model": {},
"name": "example",
"organizationId": "example",
"pd": {},
"resources": {},
"runtime": {},
"traffic": {}
},
"status": {
"conditions": [],
"createdAt": "example",
"specRevision": "example",
"status": "example",
"updatedAt": "example"
}
}
}
}Get deployment parameters
Returns the available gateways, runtime engines and versions, runtime config keys, and instance types for use when creating a deployment.
Inputs
- Gateway IDS
gateway_idslist - Optional set of gateway IDs to scope the returned parameters to. When provided, the response enumerates only the values valid across every listed gateway, so a deployment targeting this exact set and using only the returned values passes validation. Empty means all of the…
Outputs
dataobject- Response for GetDeploymentParameters
Example output
{
"data": {
"gatewayIds": [],
"resourceParameters": {
"instanceTypes": []
},
"runtimeParameters": {
"engineEnvOptions": {},
"runtimeConfigOptions": {},
"runtimeEngineOptions": {},
"runtimeVersions": {}
}
}
}Get gateway
Retrieves a single gateway by its UUID.
Inputs
- ID
idtextRequired - The unique identifier of the reservation to get, UUID format
Outputs
dataobject- Response for GetGateway
Example output
{
"data": {
"gateway": {
"spec": {
"bodyBasedRouting": {},
"coreWeaveAuth": {},
"endpointConfiguration": {},
"headerBasedRouting": {},
"id": "example",
"name": "example",
"organizationId": "example",
"pathBasedRouting": {},
"weightsAndBiasesAuth": {},
"zones": []
},
"status": {
"conditions": [],
"createdAt": "example",
"endpoints": [],
"status": "example",
"updatedAt": "example"
}
}
}
}List capacity claims
Returns every CapacityClaim that the caller can read.
Inputs
- Updated after
updated_afterdate and time - Filter by those updated after this timestamp
Outputs
dataobject- Response for ListCapacityClaims
Example output
{
"data": {
"capacityClaims": []
}
}List checkpoint hot-loads
Returns the caller's hot-loads, newest first, optionally filtered to a single deployment.
Inputs
- Parent deployment
parent_deployment_idtext - The ID of a deployment to filter by. When omitted, every hot-load in the caller's organization is returned.
- Max page size
max_page_sizenumber - The maximum number of hot-loads to return. The service may return fewer than this value. If unspecified or 0, at most 50 hot-loads are returned. The maximum value is 1000; values above 1000 are coerced to 1000. A negative value is rejected with INVALID_ARGUMENT.
- Page token
page_tokentext - A page token received from a previous ListHotLoads call. Provide this to retrieve the subsequent page. Use the same parent_deployment_id filter, including leaving it unset if it was unset in the previous call. A token used with a different filter is rejected with…
Outputs
dataobject- Response for ListHotLoads
Example output
{
"data": {
"hotLoads": [],
"nextPageToken": "example"
}
}List deployments
Returns every deployment that the caller can read, optionally filtered by parent gateway or update timestamp.
Inputs
- Parent gateway
parent_gateway_idtext - The ID of the parent gateway to filter deployments by
- Updated after
updated_afterdate and time - Filter by those updated after this timestamp
Outputs
dataobject- Response for ListDeployments
Example output
{
"data": {
"items": []
}
}List gateways
Returns every inference gateway that the caller can read, optionally filtered by update timestamp.
Inputs
- Updated after
updated_afterdate and time - Filter by those updated after this timestamp
Outputs
dataobject- Response for ListGateways
Example output
{
"data": {
"items": []
}
}List hot-load replicas
Returns the replicas of one hot-load, ordered by zone and name, in pages.
Inputs
- Hot load
hot_load_idtextRequired - The ID of the hot-load whose replicas to list
- Max page size
max_page_sizenumber - The maximum number of replicas to return. The service may return fewer than this value. If unspecified or 0, at most 50 replicas are returned. The maximum value is 1000; values above 1000 are coerced to 1000. A negative value is rejected with INVALID_ARGUMENT.
- Page token
page_tokentext - A page token received from a previous ListHotLoadReplicas call. Provide this to retrieve the subsequent page. Use the same hot_load_id and organization as the previous call; otherwise, the token is rejected with INVALID_ARGUMENT. The token is opaque and must not be…
Outputs
dataobject- Response for ListHotLoadReplicas
Example output
{
"data": {
"nextPageToken": "example",
"replicas": []
}
}Update deployment
Updates an existing deployment. The request body replaces the deployment's spec; this endpoint does not accept a field mask.
Inputs
- ID
idtextRequired - The ID of the deployment to update
- Name
nametext - The name of the deployment
- Gateway IDS
gateway_idslist - The gateways to associate the deployment with
- Runtime
runtimeobject - Runtime selection and configuration
- Resources
resourcesobject - Resource configuration for the deployment
- Model
modelobject - The model configuration
- Autoscaling
autoscalingobject - The autoscaling configuration
- Traffic
trafficobject - The traffic configuration for the deployment
- Disabled
disabledtrue/false - Disable the deployment
- Hot load
hot_loadobject - The hot-load configuration for the deployment. Immutable: it must match the value set at creation.
- Pd
pdobject - Prefill/Decode disaggregation configuration. Top-level resources and autoscaling configure the decode pool; this field configures the prefill pool. Only supported by dynamo-vllm when the pdDisaggEnabled feature flag is enabled.
Outputs
dataobject- Response for UpdateDeployment
Example output
{
"data": {
"deployment": {
"spec": {
"autoscaling": {},
"disabled": false,
"gatewayIds": [],
"hotLoad": {},
"id": "example",
"model": {},
"name": "example",
"organizationId": "example",
"pd": {},
"resources": {},
"runtime": {},
"traffic": {}
},
"status": {
"conditions": [],
"createdAt": "example",
"specRevision": "example",
"status": "example",
"updatedAt": "example"
}
}
}
}Update gateway
Updates an existing gateway. The request body replaces the gateway's spec; this endpoint does not accept a field mask.
Inputs
- ID
idtextRequired - The unique identifier of the gateway to update, UUID format
- Name
nametext - The human readable name of the gateway. Capped at 38 characters, matching CreateGatewayRequest.name.
- Zones
zoneslist - The zones to make the gateway available in, limits where deployments associated with the gateway may exist, no zones means all may be used.
- Core weave auth
core_weave_authobject - CoreWeave IAM
- Weights and biases auth
weights_and_biases_authobject - Weights & Biases
- Endpoint configuration
endpoint_configurationobject - Additional endpoint configuration options
- Body based routing
body_based_routingobject - Body based routing
- Header based routing
header_based_routingobject - Header based routing
- Path based routing
path_based_routingobject - Path based routing
Outputs
dataobject- Response for UpdateGateway
Example output
{
"data": {
"gateway": {
"spec": {
"bodyBasedRouting": {},
"coreWeaveAuth": {},
"endpointConfiguration": {},
"headerBasedRouting": {},
"id": "example",
"name": "example",
"organizationId": "example",
"pathBasedRouting": {},
"weightsAndBiasesAuth": {},
"zones": []
},
"status": {
"conditions": [],
"createdAt": "example",
"endpoints": [],
"status": "example",
"updatedAt": "example"
}
}
}
}