OVovermindlab
finetuning-jobs

List training runs — whole runs per page

The training history lists runs, not jobs: the wizard launches one job per experiment and stamps them with a shared group_id. Paging jobs would straddle a run across a page boundary and silently corrupt every per-run aggregate the client derives (status, progress, cost, duration, and the single-job test that picks the row's navigation target), so this pages over distinct runs and returns each one whole.

count is the number of matching runs. results carries one entry per run — {run_id, group_id, jobs} — where run_id is the group_id when there is one and the lone job's id otherwise, and jobs is every job of that run so client-side aggregation is always complete. Ordering applies to whole-run values before pagination. By default, the most recently created job places its run first.

Every filter of the jobs list applies (project, capability, dataset, base_model, status, provider, group_id, search).

get/api/finetuning-jobs/runs/

Request

  • Base URL
    https://api.overmindlab.ai
  • Auth
    • HTTP bearer
    • or API key in header X-Api-Key
    • or API key in cookie sessionid
    • or API key in header X-Api-Key
    • or HTTP bearer

Query parameters

base_modelstring
capabilitystring uuid
created_afterstring date-time
created_beforestring date-time
datasetstring uuid
group_idstring uuid
ordering'status' | '-status' | 'name' | '-name' | 'models' | '-models' | 'dataset' | '-dataset' | 'duration' | '-duration' | 'credits' | '-credits' | 'created_at' | '-created_at'
  • status - status
  • -status - -status
  • name - name
  • -name - -name
  • models - models
  • -models - -models
  • dataset - dataset
  • -dataset - -dataset
  • duration - duration
  • -duration - -duration
  • credits - credits
  • -credits - -credits
  • created_at - created_at
  • -created_at - -created_at
pageinteger

A page number within the paginated result set.

page_sizeinteger

Number of results to return per page.

projectstring uuid
provider'baseten' | 'modal' | 'nebius' | 'tinker' | 'together_ai'
  • together_ai - Together Ai
  • nebius - Nebius
  • tinker - Tinker
  • baseten - Baseten
  • modal - Modal
searchstring

A search term.

status'awaiting_launch' | 'cancelled' | 'deploying' | 'failed' | 'preparing' | 'preprocessing' | 'queued' | 'running' | 'submission_unknown' | 'succeeded'
  • preprocessing - Preparing data
  • awaiting_launch - Awaiting launch
  • submission_unknown - Submission Unknown
  • queued - Queued
  • preparing - Preparing
  • running - Running
  • deploying - Deploying
  • succeeded - Succeeded
  • failed - Failed
  • cancelled - Cancelled

Response

countinteger required
nextstring uri nullable
previousstring uri nullable

Example response

{
  "count": 123,
  "next": "http://api.example.org/accounts/?page=4",
  "previous": "http://api.example.org/accounts/?page=2"
}

Changes

No recorded changes to this operation across all 1 revision of this API.