> ## Documentation Index
> Fetch the complete documentation index at: https://docs.baseten.co/llms.txt
> Use this file to discover all available pages before exploring further.
# List model deployment startups
> Lists the replica startups for a model deployment.
This endpoint is in beta. The request and response structure may change before it's generally available.
Startup metrics require access for your organization. [Contact support](mailto:support@baseten.co) to request access.
## OpenAPI
````yaml get /v1/models/{model_id}/deployments/{deployment_id}/startups
openapi: 3.1.0
info:
description: REST API for management of Baseten resources
title: Baseten management API
version: 1.0.0
servers:
- url: https://api.baseten.co
security:
- BearerAuth: []
paths:
/v1/models/{model_id}/deployments/{deployment_id}/startups:
parameters:
- $ref: '#/components/parameters/model_id'
- $ref: '#/components/parameters/deployment_id'
get:
summary: Lists the startups of a model deployment
description: >-
Lists replica startups in the given time range, newest first by default.
Pass the `cursor` from a response to fetch the next page.
parameters:
- name: start_epoch_millis
in: query
required: true
description: Start of the query window, in epoch milliseconds.
schema:
title: Start Epoch Millis
type: integer
- name: end_epoch_millis
in: query
required: true
description: >-
End of the query window, in epoch milliseconds. Must be after the
start.
schema:
title: End Epoch Millis
type: integer
- name: min_duration_ms
in: query
required: false
description: Only include startups that lasted at least this many milliseconds.
schema:
anyOf:
- minimum: 0
type: integer
- type: 'null'
default: null
title: Min Duration Ms
- name: cursor
in: query
required: false
description: >-
Opaque cursor returned by a previous page. Omit to fetch the first
page. Pass the same filters and sort as the request that returned
it.
schema:
anyOf:
- type: string
- type: 'null'
default: null
title: Cursor
- name: limit
in: query
required: false
description: Maximum number of attempts to return.
schema:
default: 50
maximum: 100
minimum: 1
title: Limit
type: integer
- name: outcomes
in: query
required: false
description: >-
Only include attempts with one of these outcomes. Repeat to pass
several.
schema:
items:
$ref: '#/components/schemas/StartupOutcomeV1'
title: Outcomes
type: array
- name: phase
in: query
required: false
description: Only include attempts that spent longer than 0 ms in this phase.
schema:
anyOf:
- $ref: '#/components/schemas/StartupPhaseV1'
- type: 'null'
default: null
- name: search
in: query
required: false
description: >-
Only include attempts whose replica ID matches exactly, or whose
deployment name contains this text (case-insensitive).
schema:
anyOf:
- type: string
- type: 'null'
default: null
title: Search
- name: sort_by
in: query
required: false
description: Field to sort attempts by. Defaults to `STARTED_AT`.
schema:
anyOf:
- $ref: '#/components/schemas/StartupSortFieldV1'
- type: 'null'
default: null
- name: direction
in: query
required: false
description: Sort direction. Defaults to `desc`.
schema:
anyOf:
- $ref: '#/components/schemas/SortOrderV1'
- type: 'null'
default: null
responses:
'200':
description: ''
content:
application/json:
schema:
$ref: '#/components/schemas/StartupAttemptsV1'
x-codeSamples:
- lang: bash
source: >
curl --request GET \
--url
https://api.baseten.co/v1/models/{model_id}/deployments/{deployment_id}/startups
\
--header "Authorization: Bearer $BASETEN_API_KEY"
- lang: python
source: >-
import requests
import os
API_KEY = os.environ.get("BASETEN_API_KEY", "")
url =
"https://api.baseten.co/v1/models/{model_id}/deployments/{deployment_id}/startups"
headers = {"Authorization": f"Bearer {API_KEY}"}
response = requests.request(
"GET",
url,
headers=headers,
params={'start_epoch_millis': None, 'end_epoch_millis': None, 'min_duration_ms': None, 'cursor': None, 'phase': None, 'search': None, 'sort_by': None, 'direction': None}
)
print(response.text)
components:
parameters:
model_id:
schema:
type: string
name: model_id
in: path
required: true
deployment_id:
schema:
type: string
name: deployment_id
in: path
required: true
schemas:
StartupOutcomeV1:
description: Outcome of a whole startup attempt.
enum:
- SUCCESS
- FAILED
- STALLED
title: StartupOutcomeV1
type: string
StartupPhaseV1:
description: >-
A stage of a replica startup.
- ``SCHEDULING``: waiting for the replica to be placed on a node.
- ``IMAGE_PULL``: pulling the container image.
- ``WEIGHT_DOWNLOAD``: downloading model weights.
- ``CONTAINER_CREATE``: creating and starting the model container,
including
``CONTAINER_START`` and ``READINESS``.
- ``CONTAINER_START``: starting the model container.
- ``READINESS``: waiting for the model to pass its readiness check.
``CONTAINER_START`` and ``READINESS`` are accepted as filters but are
reported only as
part of ``CONTAINER_CREATE``.
enum:
- SCHEDULING
- IMAGE_PULL
- WEIGHT_DOWNLOAD
- CONTAINER_CREATE
- CONTAINER_START
- READINESS
title: StartupPhaseV1
type: string
StartupSortFieldV1:
description: Field to sort startup attempts by.
enum:
- OUTCOME
- STARTED_AT
- READY_AT
- DURATION
title: StartupSortFieldV1
type: string
SortOrderV1:
enum:
- asc
- desc
title: SortOrderV1
type: string
StartupAttemptsV1:
description: A page of startup attempts, newest first by default.
properties:
items:
description: Items in this page.
items:
$ref: '#/components/schemas/StartupAttemptV1'
title: Items
type: array
pagination:
$ref: '#/components/schemas/PaginationResponseV1'
description: Pagination metadata for the page.
total_count:
description: >-
Number of attempts matching the filters, counted within the 1,000
most recent startups in the window.
title: Total Count
type: integer
required:
- items
- pagination
- total_count
title: StartupAttemptsV1
type: object
StartupAttemptV1:
description: One startup attempt of a replica.
properties:
model_id:
description: Unique identifier of the model.
title: Model Id
type: string
deployment_id:
anyOf:
- type: string
- type: 'null'
description: Unique identifier of the deployment.
title: Deployment Id
deployment_name:
description: Name of the deployment.
title: Deployment Name
type: string
environment_name:
anyOf:
- type: string
- type: 'null'
description: Name of the environment. Null when querying a deployment.
title: Environment Name
replica_id:
anyOf:
- type: string
- type: 'null'
description: Identifier of the replica that started.
title: Replica Id
started_at_epoch_millis:
description: Time the attempt started, in epoch milliseconds.
title: Started At Epoch Millis
type: integer
ready_at_epoch_millis:
anyOf:
- type: integer
- type: 'null'
description: >-
Time the replica became ready, in epoch milliseconds. Null if it
never did.
title: Ready At Epoch Millis
duration_ms:
description: Duration of the attempt, in milliseconds.
title: Duration Ms
type: integer
outcome:
$ref: '#/components/schemas/StartupOutcomeV1'
description: Outcome of the attempt.
phase_summary:
description: Phases recorded during the attempt.
items:
$ref: '#/components/schemas/StartupPhaseSummaryV1'
title: Phase Summary
type: array
required:
- model_id
- deployment_id
- deployment_name
- environment_name
- replica_id
- started_at_epoch_millis
- ready_at_epoch_millis
- duration_ms
- outcome
- phase_summary
title: StartupAttemptV1
type: object
PaginationResponseV1:
properties:
has_more:
description: Whether more items exist after this page.
title: Has More
type: boolean
cursor:
anyOf:
- type: string
- type: 'null'
default: null
description: >-
Opaque cursor to pass into the next request. Null when there is no
next page.
title: Cursor
required:
- has_more
title: PaginationResponseV1
type: object
StartupPhaseSummaryV1:
description: One phase recorded during a startup attempt.
properties:
phase:
$ref: '#/components/schemas/StartupPhaseV1'
description: The phase.
status:
$ref: '#/components/schemas/StartupPhaseStatusV1'
description: Outcome of the phase.
offset_ms:
description: >-
Time from the start of the attempt to the start of the phase, in
milliseconds.
title: Offset Ms
type: integer
duration_ms:
anyOf:
- type: integer
- type: 'null'
description: >-
Duration of the phase, in milliseconds. Can be 0 for a phase that
finished within a second. Null if the phase did not finish.
title: Duration Ms
required:
- phase
- status
- offset_ms
- duration_ms
title: StartupPhaseSummaryV1
type: object
StartupPhaseStatusV1:
description: Outcome of one phase, independent of the outcome of the whole startup.
enum:
- COMPLETED
- FAILED
title: StartupPhaseStatusV1
type: string
securitySchemes:
BearerAuth:
type: http
scheme: bearer
description: >-
Send `Authorization: Bearer `. The legacy `Authorization:
Api-Key ` scheme is also accepted.
````