Build a multi-model AI video workflow

AI Video API

One endpoint family, three models. Your app picks video_model per job instead of stitching three vendors.

Why a multi-model studio API

  • One auth, one bill. Bearer nb_… key. API / Build at €59.99/mo.
  • Compare in production. Same prompt, three models, keep the winner. No new vendor contract.
  • Synchronous by default. POST returns video_url in the same response. Details in status docs.

Model picker (commercial)

API idProduct nameUse whenWatch-outs
seedance2 Seedance 2.5 Default. Social, UGC, fast iteration. Max 720p in current docs.
veo3 MiniMax H3 Cinematic, 2K, native audio. Slower than Seedance.
sora2 Sora 2 VIP 10 or 15 s storytelling. 16:9 / 9:16 only. No native audio.

Full matrices: /docs/models.php

Minimal request

POST /api/v1/text-to-video.php
curl https://nanobananavideo.com/api/v1/text-to-video.php \
  -X POST \
  -H "Authorization: Bearer nb_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A matte sneaker rotating on wet slate, slow orbit, studio light",
    "video_model": "seedance2",
    "resolution": "720p",
    "duration": 5,
    "aspect_ratio": "16:9"
  }'

Field reference and response shape: text-to-video docs.

What 300 API videos / month actually costs

The REST API is a flat €59.99 / month (API / Build plan). Generations are not billed per clip. Throughput is capped at 15 requests per 30 minutes per API key. MCP uses studio credits and does not require this plan.

155,000

Effective cost / video

€0.20

Plan (flat)

€59.99

Time at rate limit

10 h

Max throughput

720 / day

Example: 100 product videos need 7 windows of 15 jobs, about 3.5 hours if you respect Retry-After. See batch generation and rate-limit docs.

Next