Skip to main content
GET
Get model run
Use this endpoint only for runs started by POST /v1/nodes/{nodeType}/run.

Authorization

string
required
Bearer API key. Format: Bearer gx_your_api_key.

Path parameters

string
required
The runId returned by POST /v1/nodes/{nodeType} /run.

Response

string
Standalone model run ID.
string
Node type that was executed.
string | null
Sub-model used for the run, or null for single-mode models.
string
Current run status. Poll until it reaches a terminal status such as COMPLETED, FAILED, or CANCELED.
unknown
Input payload saved for the run.
unknown | null
Model output when the run has completed. Shape is model-specific.
string | null
Failure message when the run failed, otherwise null.
number
Credits charged in microcredits. This may be 0 before a run settles or when a failed run did not consume credits.
string | null
Run source, such as api, mcp, or a UI/playground source.
string
ISO timestamp when the model run was created.

Polling pattern

Poll every few seconds until the run reaches a terminal status.
Clients should keep polling the same runId until the API reports a terminal status.

Status handling

If the lookup returns 404, confirm that the runId came from the standalone model-run endpoint and belongs to the same authenticated account.

Request

Response example

Errors

Authorizations

Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Path Parameters

runId
string
required

Run ID returned by POST /v1/nodes/{nodeType}/run.

Response

Current model-run status and output.

id
string
required
nodeType
string
required
status
string
required
createdAt
string<date-time>
required
subModelId
string | null
input
any
output
any
error
string | null
userMessage
string | null

Safe, user-facing explanation of a failed run.

replacedBy
string

Present when the run failed with MODEL_RETIRED and a replacement model is known: the model ID to use instead.

creditUsed
number
source
string | null