Module evals 

The Evals API: manage evaluation runs via /evals.

![warn] This module is untested! If you encounter any issues, please report them on the repository.

An eval is a task used to measure the quality of a model’s outputs against a data source of test cases. See the OpenAI Evals guide.

Submodules: create, retrieve, update, delete, list and runs (per-eval run management, including the run output items sub-resource).

Modules

create

Create an eval via POST /evals.

delete

Delete an eval via DELETE /evals/{eval_id}.

list

List evals via GET /evals.

retrieve

Retrieve an eval via GET /evals/{eval_id}.

runs

Manage the runs of an eval via /evals/{eval_id}/runs.

update

Update an eval via POST /evals/{eval_id}.

Structs

Eval

An eval object.