Hello World

Prefer a typed client? Use the TypeScript SDK or Python SDK for typed requests, upload helpers, and built-in polling.

Get your API key

Create an API key in Settings → API Keys and export it:

export THANKYOU_API_KEY="tk_your_key_here"

Create your first generation

curl -X POST https://api.thankyouai.com/open/v1/generate \
  -H "Authorization: Bearer $THANKYOU_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/nano-banana/pro/text-to-image",
    "input": {
      "prompt": "A futuristic cityscape with flying cars, 8k",
      "aspect_ratio": "1:1"
    }
  }'

import os, time, requests

API = "https://api.thankyouai.com/open/v1"
HEADERS = {"Authorization": f"Bearer {os.environ['THANKYOU_API_KEY']}"}

# Submit
generation = requests.post(
    f"{API}/generate",
    headers={**HEADERS, "Content-Type": "application/json"},
    json={
        "model": "google/nano-banana/pro/text-to-image",
        "input": {
            "prompt": "A futuristic cityscape with flying cars, 8k",
            "aspect_ratio": "1:1",
        },
    },
).json()

# Poll until complete
while generation["status"] not in ("succeeded", "failed", "cancelled"):
    time.sleep(2)
    generation = requests.get(
        f"{API}/generations/{generation['id']}",
        headers=HEADERS,
    ).json()

print(generation["output"][0]["url"])

const API = "https://api.thankyouai.com/open/v1";
const headers = {
  Authorization: `Bearer ${process.env.THANKYOU_API_KEY}`,
  "Content-Type": "application/json",
};

// Submit
let generation = await fetch(`${API}/generate`, {
  method: "POST",
  headers,
  body: JSON.stringify({
    model: "google/nano-banana/pro/text-to-image",
    input: {
      prompt: "A futuristic cityscape with flying cars, 8k",
      aspect_ratio: "1:1",
    },
  }),
}).then((r) => r.json());

// Poll until complete
while (!["succeeded", "failed", "cancelled"].includes(generation.status)) {
  await new Promise((r) => setTimeout(r, 2000));
  generation = await fetch(`${API}/generations/${generation.id}`, {
    headers: { Authorization: headers.Authorization },
  }).then((r) => r.json());
}

console.log(generation.output[0].url);

Response

Every call returns the same generation envelope:

{
  "id": "gen_8f3k2m9x",
  "status": "succeeded",
  "model": "google/nano-banana/pro/text-to-image",
  "output": [
    {
      "url": "https://assets.thankyou.ai/gen/8f3k2m9x.png",
      "mime_type": "image/png",
      "width": 1024,
      "height": 1024
    }
  ],
  "error": null,
  "usage": {
    "credits_consumed": 0.5,
    "currency": "points"
  },
  "progress": 1.0,
  "created_at": "2026-03-29T10:00:00Z",
  "started_at": "2026-03-29T10:00:01Z",
  "finished_at": "2026-03-29T10:00:03Z"
}

output is an array — each item has a mime_type field to identify the media type, along with type-specific metadata like width, height, duration.

Long-running generations

Video and some audio workloads take longer. ThankYou returns a generation id immediately — poll the status endpoint or use webhooks.

# Step 1: submit
curl -X POST https://api.thankyouai.com/open/v1/generate \
  -H "Authorization: Bearer $THANKYOU_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "bytedance/seedance/v2/text-to-video",
    "input": {
      "prompt": "A cat playing piano in a jazz club",
      "duration": 5,
      "aspect_ratio": "16:9"
    }
  }'
# → {"id":"gen_abc123","status":"queued", ...}

# Step 2: poll lightweight status
curl https://api.thankyouai.com/open/v1/generations/gen_abc123/status \
  -H "Authorization: Bearer $THANKYOU_API_KEY"

# Step 3: fetch the full entity when done
curl https://api.thankyouai.com/open/v1/generations/gen_abc123 \
  -H "Authorization: Bearer $THANKYOU_API_KEY"

Use GET /generations/{id}/status for high-frequency polling (returns only id, status, progress). Use GET /generations/{id} when you need the full output.