Introduction
Arcads puts the leading AI video and image models in one workspace, on one credit balance. Each model's output can go straight into the next step, so an image becomes a start frame and the finished clip goes on to captions and translation.
Insight: Try the same prompt on two models before you commit to a full batch.
Video models by ad job
Ad job | Model |
Most ads, with audio, up to 30 seconds (the default) | Seedance 2.5 |
Realistic lifestyle scenes, with a start and optional end frame | Veo 3.1 |
High-fidelity product shots and complex movement | Kling 3.0 Pro |
4K output | Kling 3.0 4K or Seedance 2.0 |
Fast 1080p clips | Kling 3 Pro Turbo |
Edit a clip by describing the change, or work from image and video references | Omni Flash 1.1 |
Low-cost, stylized cinematic motion | MiniMax H3, H3 Max or H3 Max Turbo |
Clips up to 30 seconds from keyframes or image, video and audio references | Wan 3.0 or Wan 3.0 Prime |
Native audio and multilingual lip sync | Happy Horse |
Text-to-video and image-to-video from xAI | Grok Video or Grok Imagine 1.5 |
Light clips in wide or tall formats, from 21:9 to 9:16 | Seedance 2.0 Mini |
A person speaking your script | Arcads 1.0, Audio-Driven or OmniHuman 1.5 |
Image models by ad job
Ad job | Model |
Realistic images and edits (the default) | Nano-Banana Pro |
Fast results, up to 14 references, the strongest edits | Nano-Banana 2 or 2.1 |
Edits that change only what you ask, transparent backgrounds | GPT Image 2.5 Flare or Sunburst |
Prompt-following at three quality levels | GPT Image 2 |
Character and product consistency | Seedream 4.5 or Seedream 5 |
Creator-style photos | UGC Studio |
A different style, with one reference | Grok Image |
How to chain models
Step 1: Make the image
Generate the scene with an image model, or upload your product photo.
Step 2: Animate it
Click Video on the image to turn it into a clip with a video model.
Step 3: Add a person
Layer a talking actor over the clip with Merge Layers, or join them in sequence with Stitch Videos.
Step 4: Finish
Add captions, resize and translate. A workflow can run all four steps at once.
FAQs & Common Issues
Do I need separate subscriptions for each model?
No. Every model runs on your Arcads credits.