Get model deployment logs

Gets all the logs for a model deployment in the given time range, which defaults to the last 30 minutes. A failed or older deployment may only have logs from before that window; pass start_epoch_millis to widen it back to the build/deploy time.

GET /v1/models/{model_id}/deployments/{deployment_id}/logs

cURL

cURL

curl --request GET \
--url https://api.baseten.co/v1/models/{model_id}/deployments/{deployment_id}/logs \
--header "Authorization: Bearer $BASETEN_API_KEY"

200

{
  "logs": [
    {
      "timestamp": "<string>",
      "message": "<string>",
      "replica": "<string>",
      "request_id": "<string>"
    }
  ]
}

This endpoint is in beta. The request and response structure may change before it’s generally available.

Authorizations

Authorization · string · header · required

Send Authorization: Bearer <api_key>. The legacy Authorization: Api-Key <api_key> scheme is also accepted.

Path Parameters

model_id · string · required

deployment_id · string · required

Query Parameters

start_epoch_millis · integer | null

Epoch milliseconds at which to start fetching logs. Defaults to 30 minutes before the end. The window from start to end must not exceed 7 days.

end_epoch_millis · integer | null

Epoch milliseconds at which to stop fetching logs. Defaults to the current time.

direction · enum<string> | null

Sort order for logs

Available options: asc, desc

limit · integer | null · default: 500

Limit of logs to fetch in a single request

Required range: 1 <= x <= 1000

min_level · enum<string> | null

Minimum log severity to include. Omit to return all log lines, including lines that have no level. Any explicit value returns lines at or above that severity and drops lines without a level. A log severity level.

Available options: DEBUG, INFO, WARNING, ERROR

replica · string | null

Only return logs emitted by this replica (5-char short ID).

Maximum string length: 256

request_id · string | null

Only return logs tagged with this inference request ID.

Maximum string length: 256

component · string | null

Only return logs from this component.

Maximum string length: 256

search_pattern · string | null

RE2 regular expression matched against the log message. Prefer includes and excludes for plain substring matches.

Maximum string length: 256

includes · string[]

Case-sensitive substrings that must all appear in the log message.

Maximum array length: 8

excludes · string[]

Case-sensitive substrings; lines containing any of these are dropped.

Maximum array length: 8

Response

200 - application/json

A response to querying logs.

logs · LogV1 · object[] · required

Logs for a specific entity.

Show child attributes