One API key for 20 AI video models: real pricing, working code, and a $1 free tier

Description and use

Viddra’s unified API simplifies integration of 20+ AI video models with a single key, consistent request format, and per-second billing, reducing development complexity and cost management.

Problem

Developers face fragmented SDKs, auth schemes, billing, and polling loops when using multiple AI video providers, complicating integration and cost management.

Outcome

Unified API reduces integration complexity, standardizes billing, and enables cost-aware routing across multiple AI video models.

Workflow

• 1. Sign up for Viddra and obtain API key (includes $1 free credit).
• 2. Set the key in environment variable VIDDRA_API_KEY.
• 3. Send a POST request to https://api.viddra.com/v1/video/generations with JSON body containing model, prompt, and duration.
• 4. Receive 202 Accepted and a task_id.
• 5. Poll the task status every 10 seconds using GET https://api.viddra.com/v1/video/generations/{task_id} until status is succeeded or failed.
• 6. On success, retrieve video_url; on failure, error is returned and cost is refunded automatically.
• 7. Optionally switch the model field to route by price/quality (e.g., kling3.0, wan3.0-video, hailuo-h3).

Evidence

One endpoint handles every model. Auth is a standard bearer key:

Source

Billing works on hold-then-settle: the estimated cost is held when the task starts, and on failure the hold is refunded automatically

Source

$1 free credit on signup — enough to test 5–8 real generations across models, no card required

Source