QuickstartMarkdown

Price it first

POST /v1/generations/quote takes the same body as a submit and answers with what it would cost. It is free: nothing is created and nothing is held. Quote the exact request, then pass the quote as max_cost_usd.

curl -s -X POST https://vidlab.ai/v1/generations/quote \
  -H "Authorization: Bearer $VIDLAB_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.5",
    "input": { "resolution": "720p", "duration": 5, "aspect_ratio": "9:16" }
  }'

import os, requests

r = requests.post(
    "https://vidlab.ai/v1/generations/quote",
    headers={"Authorization": f"Bearer {os.environ['VIDLAB_API_KEY']}"},
    json={
        "model": "seedance-2.5",
        "input": {"resolution": "720p", "duration": 5, "aspect_ratio": "9:16"},
    },
)
quote = r.json()["data"]
print(quote["quoted_usd"], quote["enough_credit"])

const res = await fetch("https://vidlab.ai/v1/generations/quote", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.VIDLAB_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "seedance-2.5",
    input: { resolution: "720p", duration: 5, aspect_ratio: "9:16" },
  }),
});
const { data: quote } = await res.json();
console.log(quote.quoted_usd, quote.enough_credit);

json

{
  "data": {
    "object": "quote",
    "model": "seedance-2.5",
    "variant": "720p",
    "unit": "per_second",
    "units": 5,
    "price_usd": 1.05,
    "quoted_usd": 1.05,
    "reserved_usd": 1.05,
    "estimated_seconds": 90,
    "available_usd": 25,
    "enough_credit": true
  }
}

Field	Type	What it is
`quoted_usd`	number	What a success is charged.
`reserved_usd`	number	What a submit would hold on your balance while it runs. Released if it fails.
`estimated_seconds`	number	How long the model usually takes. Compare it with `max_wait_s`.
`available_usd`	number	What your balance can spend now.
`enough_credit`	boolean	Whether `available_usd` covers `reserved_usd`.
`variant`, `unit`, `units`		The priced option, what one unit is, and how many units the request counts. See Model list.

Content fields (prompt, images, video) may be left out: they do not change the price. A request the model would refuse answers the same 400 or 422 as a submit.

No key yet

POST /v1/estimate answers the same price without a key and without your balance. Free, and rate limited per address.

curl -s -X POST https://vidlab.ai/v1/estimate \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-image-2", "input": {"resolution": "1k", "n": 1}}'

json

{ "data": { "model": "gpt-image-2", "variant": "1k", "unit": "per_image", "units": 1, "price_usd": 0.03 } }