Update Experiment
Path parameters
Request
Workspace-unique experiment name.
Human-readable purpose of the experiment.
Human- or agent-authored summary of the experiment’s findings.
Free-form producer metadata for the experiment.
Default sort for this experiment’s evaluations list, as a sort-param string: a comma-separated, ordered list of fields where the first is the primary sort and the rest break ties (leading ’-’ on a field = descending), e.g. ‘-evaluators.reward.mean,cost_usd.mean’. Defaults to ‘-created_at’. Accepts any field the evaluations list sort param does; clients apply it as the list sort param.
Default X/Y metrics for the experiment’s Pareto view. Omit to preserve the existing value on update; on create, defaults to cost vs. latency.
Saved column order and column visibility for the experiment’s evaluations table. Omit to leave the saved layout unchanged; on create that means no layout is saved. An empty order and hidden pair is itself a saved layout — one that shows every column — and is distinct from having none.
Whether this Experiment is marked as a favorite. Defaults to false on create; omit on update to preserve the existing value.
Whether Studio should display this Experiment’s Evaluation results over time. Defaults to false on create; omit on update to preserve the existing value.
Response
Default X/Y metrics for a group’s cost-vs-accuracy Pareto view.
Metric ids use the same vocabulary as the evaluations list sort/filter fields — cost_usd,
latency_ms, tokens, or evaluators.<name>. Defaults to cost (x) vs latency (y): both exist for
every group, so the chart always has something to render before anyone customizes it.
Number of live (non-soft-deleted) evaluations in this experiment.