Launch a machine
launch_machineRequest a cost quote, then launch a GPU machine after approval; booked hours are charged upfront and unused time is refunded.
Instructions
Launch a GPU machine. Costs money: call first with confirm=false to get a quote (nothing happens), show it to the person, and only after they agree call again with the same arguments, confirm=true and expected_total_usd. The booked time is charged up front from the team's balance.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"gpu": "<gpu>",
"app": "<app>",
"hours": "<hours>"
},
"name": "launch_machine"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| app | Yes | App from list_apps, e.g. datascience | |
| gpu | Yes | GPU slug from list_gpus, e.g. rtx-a6000 | |
| hours | Yes | How long to book; unused time is refunded | |
| label | No | A name for the machine | |
| model | No | For apps with models; left out = our recommendation for this GPU | |
| region | No | Where it must run | any |
| confirm | No | false = only quote (default); true = launch, after the person agreed to the quote | |
| project | No | Project name or id to restore into ~/work and save back | |
| endpoint | No | Endpoint name or id this machine serves | |
| gpu_count | No | ||
| expected_total_usd | No | With confirm=true: the quote's total_usd the person agreed to | |
| keep_running_hours | No | Renew this many hours at a time until stopped |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| quote | No | ||
| status | Yes | quote = nothing happened yet; done = it was carried out | |
| machine | No |