# Eval Run

A benchmark evaluation run. The public benchmark API currently evaluates chat-mode responses.

## The GenieEvalRunResponse object

- [eval-run/object.md](eval-run/object.md)

## Endpoints

### Get a benchmark evaluation run

- [GET /api/2.0/genie/spaces/{space_id}/eval-runs/{eval_run_id}](genie-get-eval-run.md)

### List benchmark evaluation runs

- [GET /api/2.0/genie/spaces/{space_id}/eval-runs](genie-list-eval-runs.md)

### Start a benchmark evaluation run

- [POST /api/2.0/genie/spaces/{space_id}/eval-runs](genie-create-eval-run.md)

### Get benchmark evaluation result details

- [GET /api/2.0/genie/spaces/{space_id}/eval-runs/{eval_run_id}/results/{result_id}](genie-get-eval-result-details.md)

### List results for a benchmark evaluation run

- [GET /api/2.0/genie/spaces/{space_id}/eval-runs/{eval_run_id}/results](genie-list-eval-results.md)