Run a batch of cases through an agent or swarm and grade them

The whole measure loop in one call: attach evals, run the cases, return the per-case, per-agent, per-step graded results.

This endpoint expects an object.

targetobjectRequired

caseslist of objectsRequired

evalslist of objectsOptional

run_typeenumOptional

Allowed values:

scoring_modeenumOptional

Allowed values:

labelstringOptional

Simulation complete

401

Unauthorized Error

$	curl -X POST https://api.recursiv.io/api/v1/evidence/simulate \
>	-H "Authorization: Bearer <token>" \
>	-H "Content-Type: application/json" \
>	-d '{
>	"target": {
>	"type": "agent",
>	"id": "3fa85f64-5717-4562-b3fc-2c963f66afa6"
>	},
>	"cases": [
>	{
>	"input": "Translate the following English sentence to French: '\''Hello, how are you?'\''"
>	}
>	]
>	}'