Developer API

API Documentation

Generate AI videos programmatically. Same models, same quality, your codebase.

Quick Start

cURL

# 1. Start a generation
curl -X POST https://www.bananapro.video/api/generate \
  -H "Authorization: Bearer bpv_live_xxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A cat astronaut floating in space",
    "model": "seedance"
  }'

# 2. Poll for the result
curl "https://www.bananapro.video/api/generation-status?id=YOUR_ID" \
  -H "Authorization: Bearer bpv_live_xxxxxxxxxxxx"

Authentication

All API requests require an API key. Include it in the Authorization header.

Authorization: Bearer bpv_live_xxxxxxxxxxxx

Generate Video

POST/api/generate

Starts a new video generation. The video is generated asynchronously — use polling or a webhook callback to get the result.

Request Body

prompt

stringRequired

Text description of the video to generate

model

string

Model ID (see Available Models below, default: seedance)

aspectRatio

string

16:9, 9:16, 1:1, 4:3, 3:4, 21:9 (default: 16:9)

duration

number

Video duration in seconds (model-dependent)

resolution

string

480p, 720p, 1080p (model-dependent)

imageUrl

string

URL of an image for image-to-video generation

callback_url

string

URL to receive a POST when the video is ready (optional, alternative to polling)

Response

{
  "success": true,
  "data": {
    "id": "fa6fd0ef-f5a8-44af-b8f4-565f0521f9cd",
    "status": "processing",
    "model": "seedance",
    "prompt": "A cat astronaut floating in space",
    "created_at": "2026-03-27T21:42:05.788Z"
  },
  "remaining_credits": 1450
}

Check Status

GET/api/generation-status?id=YOUR_ID

Poll this endpoint to check the status of a generation. Use the id returned by the generate endpoint.

While processing

{
  "id": "fa6fd0ef-...",
  "status": "processing",
  "video_url": null,
  "thumbnail_url": null,
  "created_at": "2026-03-27T21:42:05.788Z"
}

When completed

{
  "id": "fa6fd0ef-...",
  "status": "completed",
  "video_url": "https://fal.media/files/.../video.mp4",
  "thumbnail_url": null,
  "created_at": "2026-03-27T21:42:05.788Z"
}

Poll every 5-10 seconds. Generation typically takes 1-3 minutes depending on the model.

Webhook Callback

Instead of polling, you can provide a callback_url when creating a generation. We'll POST the result to your URL when the video is ready — no polling needed.

Request with callback

curl -X POST https://www.bananapro.video/api/generate \
  -H "Authorization: Bearer bpv_live_xxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A cat astronaut floating in space",
    "model": "seedance",
    "callback_url": "https://your-app.com/webhook/video-ready"
  }'

Callback payload

When the video is ready (or fails), we send a POST request to your callback_url:

// POST to your callback_url
{
  "id": "fa6fd0ef-...",
  "status": "completed",
  "video_url": "https://fal.media/files/.../video.mp4",
  "thumbnail_url": null,
  "created_at": "2026-03-27T21:42:05.788Z",
  "completed_at": "2026-03-27T21:43:12.673Z"
}

Available Models

sora-2Sora 2

OpenAI's flagship model - cinematic quality

10 CR/s

veo-3.1Veo 3.1 Fast

Google's model with native audio

20 CR/s

kling-2.6Kling 2.6 Pro

High-fidelity with audio support

10 CR/s

kling-1.6-elementsKling Elements

Multi-image fusion (2-4 images)

10 CR/s

wan-2.5Wan 2.5

Ultra-realistic, best value

5 CR/s

ray-3Ray 2 Flash

Luma's fast creative generation

10 CR/s

seedanceSeedance 1.0 Lite

ByteDance - fast & efficient

5 CR/s

minimaxHailuo 02 Standard

MiniMax - ultra-realistic humans

10 CR/s

Code Examples

Python (with polling)

import requests
import time

API_KEY = "bpv_live_xxxxxxxxxxxx"
BASE_URL = "https://www.bananapro.video"
HEADERS = {
    "Authorization": f"Bearer {API_KEY}",
    "Content-Type": "application/json"
}

# 1. Start generation
response = requests.post(f"{BASE_URL}/api/generate", headers=HEADERS, json={
    "prompt": "A cat astronaut floating in space",
    "model": "seedance",
    "aspectRatio": "16:9"
})

generation = response.json()
gen_id = generation["data"]["id"]
print(f"Generation started: {gen_id}")
print(f"Remaining credits: {generation['remaining_credits']}")

# 2. Poll until completed
while True:
    status = requests.get(
        f"{BASE_URL}/api/generation-status?id={gen_id}",
        headers=HEADERS
    ).json()

    if status["status"] == "completed":
        print(f"Video ready: {status['video_url']}")
        break
    elif status["status"] == "failed":
        print(f"Generation failed: {status.get('error')}")
        break

    time.sleep(5)

JavaScript / Node.js (with polling)

const API_KEY = "bpv_live_xxxxxxxxxxxx";
const BASE_URL = "https://www.bananapro.video";

// 1. Start generation
const response = await fetch(`${BASE_URL}/api/generate`, {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${API_KEY}`,
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    prompt: "A cat astronaut floating in space",
    model: "seedance",
    aspectRatio: "16:9"
  })
});

const { data, remaining_credits } = await response.json();
console.log(`Generation started: ${data.id}`);
console.log(`Remaining credits: ${remaining_credits}`);

// 2. Poll until completed
while (true) {
  const status = await fetch(
    `${BASE_URL}/api/generation-status?id=${data.id}`,
    { headers: { "Authorization": `Bearer ${API_KEY}` } }
  ).then(r => r.json());

  if (status.status === "completed") {
    console.log(`Video ready: ${status.video_url}`);
    break;
  } else if (status.status === "failed") {
    console.error(`Failed: ${status.error}`);
    break;
  }

  await new Promise(r => setTimeout(r, 5000));
}

Rate Limits & Billing

20 generations per hour per API key
Credits are deducted per generation based on model and duration
Failed generations are automatically refunded
API access requires a paid plan

Banana Pro Video

Developer API

API Documentation

Generate AI videos programmatically. Same models, same quality, your codebase.

Quick Start

cURL

# 1. Start a generation
curl -X POST https://www.bananapro.video/api/generate \
  -H "Authorization: Bearer bpv_live_xxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A cat astronaut floating in space",
    "model": "seedance"
  }'

# 2. Poll for the result
curl "https://www.bananapro.video/api/generation-status?id=YOUR_ID" \
  -H "Authorization: Bearer bpv_live_xxxxxxxxxxxx"

Authentication

All API requests require an API key. Include it in the Authorization header.

Authorization: Bearer bpv_live_xxxxxxxxxxxx

Generate Video

POST/api/generate

Starts a new video generation. The video is generated asynchronously — use polling or a webhook callback to get the result.

Request Body

prompt

stringRequired

Text description of the video to generate

model

string

Model ID (see Available Models below, default: seedance)

aspectRatio

string

16:9, 9:16, 1:1, 4:3, 3:4, 21:9 (default: 16:9)

duration

number

Video duration in seconds (model-dependent)

resolution

string

480p, 720p, 1080p (model-dependent)

imageUrl

string

URL of an image for image-to-video generation

callback_url

string

URL to receive a POST when the video is ready (optional, alternative to polling)

Response

{
  "success": true,
  "data": {
    "id": "fa6fd0ef-f5a8-44af-b8f4-565f0521f9cd",
    "status": "processing",
    "model": "seedance",
    "prompt": "A cat astronaut floating in space",
    "created_at": "2026-03-27T21:42:05.788Z"
  },
  "remaining_credits": 1450
}

Check Status

GET/api/generation-status?id=YOUR_ID

Poll this endpoint to check the status of a generation. Use the id returned by the generate endpoint.

While processing

{
  "id": "fa6fd0ef-...",
  "status": "processing",
  "video_url": null,
  "thumbnail_url": null,
  "created_at": "2026-03-27T21:42:05.788Z"
}

When completed

{
  "id": "fa6fd0ef-...",
  "status": "completed",
  "video_url": "https://fal.media/files/.../video.mp4",
  "thumbnail_url": null,
  "created_at": "2026-03-27T21:42:05.788Z"
}

Poll every 5-10 seconds. Generation typically takes 1-3 minutes depending on the model.

Webhook Callback

Instead of polling, you can provide a callback_url when creating a generation. We'll POST the result to your URL when the video is ready — no polling needed.

Request with callback

curl -X POST https://www.bananapro.video/api/generate \
  -H "Authorization: Bearer bpv_live_xxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A cat astronaut floating in space",
    "model": "seedance",
    "callback_url": "https://your-app.com/webhook/video-ready"
  }'

Callback payload

When the video is ready (or fails), we send a POST request to your callback_url:

// POST to your callback_url
{
  "id": "fa6fd0ef-...",
  "status": "completed",
  "video_url": "https://fal.media/files/.../video.mp4",
  "thumbnail_url": null,
  "created_at": "2026-03-27T21:42:05.788Z",
  "completed_at": "2026-03-27T21:43:12.673Z"
}

Available Models

sora-2Sora 2

OpenAI's flagship model - cinematic quality

10 CR/s

veo-3.1Veo 3.1 Fast

Google's model with native audio

20 CR/s

kling-2.6Kling 2.6 Pro

High-fidelity with audio support

10 CR/s

kling-1.6-elementsKling Elements

Multi-image fusion (2-4 images)

10 CR/s

wan-2.5Wan 2.5

Ultra-realistic, best value

5 CR/s

ray-3Ray 2 Flash

Luma's fast creative generation

10 CR/s

seedanceSeedance 1.0 Lite

ByteDance - fast & efficient

5 CR/s

minimaxHailuo 02 Standard

MiniMax - ultra-realistic humans

10 CR/s

Code Examples

Python (with polling)

import requests
import time

API_KEY = "bpv_live_xxxxxxxxxxxx"
BASE_URL = "https://www.bananapro.video"
HEADERS = {
    "Authorization": f"Bearer {API_KEY}",
    "Content-Type": "application/json"
}

# 1. Start generation
response = requests.post(f"{BASE_URL}/api/generate", headers=HEADERS, json={
    "prompt": "A cat astronaut floating in space",
    "model": "seedance",
    "aspectRatio": "16:9"
})

generation = response.json()
gen_id = generation["data"]["id"]
print(f"Generation started: {gen_id}")
print(f"Remaining credits: {generation['remaining_credits']}")

# 2. Poll until completed
while True:
    status = requests.get(
        f"{BASE_URL}/api/generation-status?id={gen_id}",
        headers=HEADERS
    ).json()

    if status["status"] == "completed":
        print(f"Video ready: {status['video_url']}")
        break
    elif status["status"] == "failed":
        print(f"Generation failed: {status.get('error')}")
        break

    time.sleep(5)

JavaScript / Node.js (with polling)

const API_KEY = "bpv_live_xxxxxxxxxxxx";
const BASE_URL = "https://www.bananapro.video";

// 1. Start generation
const response = await fetch(`${BASE_URL}/api/generate`, {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${API_KEY}`,
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    prompt: "A cat astronaut floating in space",
    model: "seedance",
    aspectRatio: "16:9"
  })
});

const { data, remaining_credits } = await response.json();
console.log(`Generation started: ${data.id}`);
console.log(`Remaining credits: ${remaining_credits}`);

// 2. Poll until completed
while (true) {
  const status = await fetch(
    `${BASE_URL}/api/generation-status?id=${data.id}`,
    { headers: { "Authorization": `Bearer ${API_KEY}` } }
  ).then(r => r.json());

  if (status.status === "completed") {
    console.log(`Video ready: ${status.video_url}`);
    break;
  } else if (status.status === "failed") {
    console.error(`Failed: ${status.error}`);
    break;
  }

  await new Promise(r => setTimeout(r, 5000));
}

Rate Limits & Billing

20 generations per hour per API key
Credits are deducted per generation based on model and duration
Failed generations are automatically refunded
API access requires a paid plan