Skip to main content
POST
/
v1
/
rentals
/
preview
curl -X POST https://gpuoutlet.ai/v1/rentals/preview \
  -H "Content-Type: application/json" \
  --cookie "sid=$SID" \
  -d '{"familyId":"h100-80gb","gpuCount":1}'

{
  "pricePerHourCents": 280,
  "estimatedProvisionSeconds": 45,
  "matchedListing": {
    "id": "lst_01H...",
    "providerCode": "vast",
    "region": "US-East",
    "model": "H100_SXM_80GB"
  }
}
Useful for showing the user the price before they commit. Doesn’t reserve capacity, doesn’t debit the wallet.
​
familyId
string
required
GPU family id, e.g. rtx-4090, h100-80gb, b200. See the Pricing reference for the full list.
​
gpuCount
number
required
1, 2, 4, or 8.
​
templateId
string
Optional. e.g. ubuntu-24-04, pytorch, vllm, ollama, a1111, text-gen-webui. Defaults to ubuntu-24-04.
​
region
string
Optional region preference. If omitted, the cheapest listing wins.
​
pricePerHourCents
number
Total hourly rate for the matched listing.
​
estimatedProvisionSeconds
number
How long boot typically takes for this combination — usually 20–60.
​
matchedListing
object
The specific listing we’d use if you called POST /v1/rentals next.
curl -X POST https://gpuoutlet.ai/v1/rentals/preview \
  -H "Content-Type: application/json" \
  --cookie "sid=$SID" \
  -d '{"familyId":"h100-80gb","gpuCount":1}'

{
  "pricePerHourCents": 280,
  "estimatedProvisionSeconds": 45,
  "matchedListing": {
    "id": "lst_01H...",
    "providerCode": "vast",
    "region": "US-East",
    "model": "H100_SXM_80GB"
  }
}