Sign in
EVALS

Run Eval

Run an evaluation for a test case. This simulates the conversation by calling the /ask endpoint and evaluates responses.

post/evals/run

Query parameters

navigationSourcestring nullable

Request body

test_case_idstring required
session_idstring nullable

Response

Successful Response

idstring required
test_case_idstring nullable
test_case_namestring required
test_case_descriptionstring nullable
test_case_tagsstring[] nullable
statusstring nullable
overall_scorenumber nullable
judge_errorsinteger nullable
judge_modelstring nullable
created_atstring required
session_idstring nullable
run_namestring nullable
tenantstring
run_tenantstring nullable
trace_idstring nullable
datamate_ids_usedstring[] nullable
knowledge_base_ids_usedinteger[] nullable
tools_invokedobject nullable
knowledge_bases_queriedinteger[] nullable

Changes

Changed in 1 of the 28 revisions of this API.25

  • 103580dad81625See the full diff
    • ▲

      the response property overall_score became optional for the status 200

      response-property-became-optional

    • ▲

      response property overall_score list-of-types was widened by adding types null to media type application/json of response 200

      response-property-list-of-types-widened

    • ○

      added the optional property judge_errors to the response with the 200 status

      response-optional-property-added

    • ○

      added the optional property judge_model to the response with the 200 status

      response-optional-property-added

    • ○

      added the optional property run_tenant to the response with the 200 status

      response-optional-property-added

    • ○

      added the optional property status to the response with the 200 status

      response-optional-property-added

    • ○

      added the optional property test_case_id to the response with the 200 status

      response-optional-property-added