MODELS
One wallet, many models.
Pick the right model per shot — speed, quality, or native audio — without juggling separate accounts or credit balances.
Supported models
| Model | Modalities | Max duration | Resolution | Native audio | Cost | |
|---|---|---|---|---|---|---|
Seedance 2.5 ByteDance | Text-to-video · Image-to-video | 10s | Up to 1080p | Yes | 10 credits / 5s | View model → |
Veo 3 Google DeepMind | Text-to-video · Image-to-video | 8s | Up to 4K | Yes | 16 credits / 5s | View model → |
Kling 2.5 Kuaishou | Image-to-video | 10s | Up to 1080p | No | 14 credits / 5s | View model → |
Runway Gen-4 Runway | Video-to-video · Image-to-video | 6s | Up to 1080p | No | 18 credits / 5s | View model → |
Placeholder catalog — wires to the real models table (blueprint §12) once provider adapters are live.
SC 02 · PROOF OF GENERATION
Synthesized by prompt. Framed for cinema.
Every shot below was generated directly from a single natural-language prompt with specified lens optics, camera motion, and continuous style locks. Click any shot to expand it.
