The AI models you can generate with
Scenes AI runs on models from Google, ByteDance, Kuaishou, MiniMax and Suno. You choose one by what it is good at — visual quality, price, or how long a single scene can run — and everything else, including which reference images it receives, is handled for you. Prices below are what you are actually charged; one credit is one US dollar.
Video engines
These animate each generated still into a clip. Quality scores come from measured image-to-video preference testing with audio excluded, so they compare the picture rather than the soundtrack.
- Veo 3.1 Litefrom $0.20 per clip
by Google
Fastest and cheapest · ideal for iterating.
- Quality
- 62/100
- Max clip
- 8s
- References
- 3
- Veo 3.1 Fastfrom $0.39 per clip
by Google
Balanced quality and cost · strongest in-video audio.
- Quality
- 68/100
- Max clip
- 8s
- References
- 3
- Kling 3.0from $0.091 per second
by Kuaishou
Native multi-shot (up to 5 shots) · character binding with voice · up to 15s.
- Quality
- Not rated
- Max clip
- 15s
- References
- 12
- Seedance 2.0from $0.12 per second
by ByteDance
Best visual quality · multi-shot scenes up to 15s · 9 reference images.
- Quality
- 100/100
- Max clip
- 15s
- References
- 9
- MiniMax H3from $0.15 per second
by MiniMax
Omni-reference (9 images + video + audio) · native audio · up to 15s.
- Quality
- Not rated
- Max clip
- 15s
- References
- 9
- Seedance 2.5from $0.18 per second
by ByteDance
Longest single shot — up to 30s unbroken · 30 reference images · 720p.
- Quality
- Not rated
- Max clip
- 30s
- References
- 30
Image and music engines
Stills carry your character sheets and location plates; music is written for your video rather than picked from a library.
- Nano Banana Profrom $0.12 per image
by Google
Images · 1K, 2K, 4K
- Nano Banana 2from $0.052 per image
by Google
Images · 1K, 2K, 4K
- Suno V5.5from $0.078 per track
by Suno
Original music, with lyrics or instrumental
Compare engines head-to-head
The same specs above, side by side for two engines at a time.