Pick the right length
DomoAI suggests 5 s for strongly stylized images or clear emotions, and 10 s for prominent human subjects and narrative action, because frame interpolation slows motion down.
Domo AI · Model notes
Domo AI v2.3
The image-to-video model, what it improved, and what replaced it
v2.3 was the generation of DomoAI’s image-to-video model with two modes, Faster for quick results and Advanced for more precise animation. It has since been succeeded by v2.4, which is the version DomoAI’s help center and API list today. This page explains what v2.3 was known for, how it compares, and which current model to use instead.
The basics
What was Domo AI v2.3?
v2.3 was an update to DomoAI’s image-to-video engine, the model that animates a still picture from a prompt or a template. Creator reviews described it as faster and more accurate than earlier versions. It was offered in two modes so you could trade speed for precision.
FasterQuick results for testing ideas, social clips and templates
AdvancedMore careful motion and detail for busier, more complex scenes
Users and reviewers mainly credited v2.3 with smoother motion, better handling of detailed scenes, and more natural facial movement. DomoAI hasn’t published official benchmarks for these.
Version timeline
How DomoAI’s models evolved
Only dated, sourced milestones. Exact release dates for some 2.x versions weren’t published.
- Mar 2024Video 2.0 style upgrade
DomoAI upgraded its styles to video 2.0, including Japanese Anime 2.1, Paper Art, Pixel Art, Flat Color Anime, Chinese Ink Painting and Detailed Anime.
- Aug 2024Anime 5.3
A new anime model, as reported by creators testing it on release.
- Mar 2025Image-to-video upgrade
Better stability and speed, much stronger prompt understanding, and more detail kept in complex scenes, per AIbase.
- 2.xv2.3 · Faster & Advanced
Image-to-video with separate speed and precision modes.
- Nowv2.4 · Faster & Advanced
The current image-to-video and text-to-video models in the app and the API, at 5 or 10 s per generation.
- 2026Seedance 2.0 in Text to Video
4–15 s clips, 480P–1080P, optional native audio.
- Aug 2026Omni Reference
Seedance 2.5 with up to 50 references, later joined by MiniMax H3.
Comparison
v2.3 vs v2.4: what’s documented
| Aspect | v2.3 | v2.4 (current) |
|---|---|---|
| Modes | Faster · Advanced | Faster (speed) · Advanced (precision) |
| Used in | Image to Video, templates | Image to Video, Text to Video, API |
| Clip length | Short clips | 5 or 10 s per generation |
| Relax mode | — | Standard plan and above |
| API model IDs | Not listed in current docs | animate-2.4-* · t2v-2.4-* |
| API cost | — | Faster 2 credits/s · Advanced 5 credits/s ($0.02 each) |
| Availability | Succeeded | Current |
“—” means DomoAI hasn’t published it. We don’t guess credit costs or quality numbers for older versions.
Interactive
Used v2.3? Find the model to use now
Tell us how you used v2.3 and what you need, and we’ll point you to the current option.
I used v2.3 for…
I mostly chose…
Now I also need…
Use this now
For developers
Migrating API code to 2.4
The current API docs list 2.4 model IDs. If your integration still points at an older model name, update it and test with a short clip.
- Image to video:
animate-2.4-fasteroranimate-2.4-advanced - Text to video:
t2v-2.4-fasterort2v-2.4-advanced - Seconds: 1–10 per request
- Faster costs 2 credits/s; Advanced costs 5 credits/s
Before → after
"model": "<older model id>"
"model": "animate-2.4-faster"
POST https://api.domoai.com/v1/video/image2video
{ "model": "animate-2.4-faster",
"image": { "bytes_base64_encoded": "…" },
"prompt": "character waves at the camera",
"seconds": 5 }Check the official API docs ↗ for the exact model list before shipping.
Better results
Image to Video guide →Tips for DomoAI’s 2.x image-to-video models
Faster first, then Advanced
Find the motion and prompt on Faster, then re-render the keeper on Advanced for precision.
Clean input, clear prompt
A sharp image with one clear subject, plus one action and one camera move in the prompt, gives the most stable results.
Prompt guide →What is Domo AI v2.3?
v2.3 was a generation of DomoAI’s image-to-video model, offered in Faster and Advanced modes so you could choose between speed and precision. It has since been succeeded by v2.4.
Is v2.3 still available?
DomoAI’s help center and API documentation now list v2.4 Faster and v2.4 Advanced as the current image-to-video models. Older versions may no longer be selectable, so use v2.4 for the same workflow.
What’s the difference between Faster and Advanced?
Faster is tuned for speed, which makes it good for testing ideas and quick social clips. Advanced is tuned for precision, which suits complex scenes and final renders. Through the API, Faster costs 2 credits per second and Advanced 5.
What did v2.3 improve?
Creators reported smoother motion, better handling of detailed and busy scenes, and more natural facial movement than earlier versions. DomoAI didn’t publish official benchmark numbers.
How long are clips with the current 2.4 models?
5 or 10 seconds per generation. DomoAI suggests 5 s for strongly stylized images and 10 s for human subjects and narrative action. For up to 15 s with audio, use Seedance 2.0 in Text to Video.
Which model should I use instead of v2.3?
v2.4 Faster if you used v2.3 Faster, and v2.4 Advanced if you used Advanced. For longer clips or audio, use Seedance 2.0. For consistent characters across shots, use Omni Reference.
Do I need to change my API code?
If your code uses an older model ID, switch to the current 2.4 IDs, such as animate-2.4-faster or t2v-2.4-faster, and check the official API docs for the latest list.
Same workflow, newer model.
Animate your images with DomoAI’s current 2.4 models. Start with free credits.