v2
latestOpenAPI 3.1.02026-08-05267431678.1 KBList Evaluations
List all evaluations for the authenticated user.
Returns model and benchmark evaluations with their metrics. Used by the frontend to build the evaluations matrix view.
If project_id is the sentinel "default", returns only evaluations whose project_id is null (the unallocated bucket). If project_id is any other value, it must be a valid UUID and the response includes evaluations for that project plus orphan evaluations (legacy behavior). If model_id is provided, only evaluations for that model are returned.
Query parameters
Filter by project ID. Pass the literal value 'default' to return only orphan evaluations (project_id IS NULL). For real project UUIDs the response also includes orphans for backward compatibility.
Filter by project ID. Pass the literal value 'default' to return only orphan evaluations (project_id IS NULL). For real project UUIDs the response also includes orphans for backward compatibility.
Filter evaluations by the evaluated model's ID
Filter evaluations by the evaluated model's ID
Maximum evaluations to return in this page.
Maximum evaluations to return in this page.
Opaque cursor returned by the previous page.
Opaque cursor returned by the previous page.
Response
Successful Response