AI Models for Images and Video
All the AI models we run, in one place: 16 video models and 12 image models, from Seedance 2.0, Veo 3.1 and Kling 3.0 to Nano Banana Pro and GPT Image 2. Open any model page to try it right away, with the credits shown before you run.
AI Video Models
AI video models with their own page. Start from a prompt or a photo, choose the length and resolution, and add sound where the model offers it.
New
ByteDanceSeedance 2.0ByteDance's flagship: film-like camera work, big action and physics, sound at no extra cost. Does not take photos of real people.62 credits per 5 s clip
New
KuaishouKling 3.0Kling's top tier in 1080p: smooth, controlled motion; start and end frames.23 credits per 5 s clip
New
GoogleVeo 3.1Google's full Veo: rich detail and natural light; start and end frames.33 credits per 4 s clip
New
xAIGrok ImaginexAI's Grok video: lively motion, with sound.29 credits per 5 s clip
MiniMaxHailuo 2.3Strong push-ins and lively faces; takes a few minutes.12 credits per 6 s clipAI Image Models
Image models with their own page. Every one makes images from text, and the ones that take reference images can also edit your photos.
GoogleNano Banana ProBest at complex scenes and text inside images. Up to 4K.From 7 credits per image
GoogleNano Banana 2Close to Pro quality, faster and cheaper. 1K or 2K.From 4 credits per image
GoogleNano BananaQuick everyday images and edits at a low price.From 2 credits per image
OpenAIGPT Image 2Follows prompts precisely; clean typography.From 3 credits per image
IdeogramIdeogram 3.0Strong at posters, logos and words inside images.From 3 credits per imageEvery AI Model on ImageToPsd.AI
All the models in our composer, what each one is best at, and its credits. Models with their own page link to it, and every model is one click away in the model picker.
| Model | Best for | Credits |
|---|---|---|
| Seedance 2.5 | ByteDance's newest: up to 30 seconds in one continuous take, sound at no extra cost. Does not take photos of real people. | 96 |
| Seedance 2.0 | ByteDance's flagship: film-like camera work, big action and physics, sound at no extra cost. Does not take photos of real people. | 62 |
| Seedance 1.5 Pro | Follows camera directions closely and keeps faces and products intact. | 6 |
| Veo 3.1 | Google's full Veo: rich detail and natural light; start and end frames. | 33 |
| Veo 3.1 Fast | Natural motion and light from Google; start and end frames. | 17 |
| Veo 3.1 Lite | Google’s lower-cost Veo: calm, steady motion. | 5 |
 Gemini Omni Flash 1.1 | Google's newest video model: real-world physics, with sound. | 21 |
| Kling 3.0 Pro | Kling's top tier in 1080p: smooth, controlled motion; start and end frames. | 23 |
| Kling 3.0 Turbo | Quick clips with smooth, steady camera moves. | 23 |
| MiniMax H3 Max | Ranked first for image to video in blind tests (October 2026), with sound. | 19 |
| Hailuo 2.3 Fast | Brings a photo to life quickly. | 8 |
| Hailuo 2.3 | Strong push-ins and lively faces; takes a few minutes. | 12 |
 HappyHorse 1.1 | Alibaba's flagship in up to 1080p, with sound and lip sync. | 29 |
| Wan 3.0 | Bold camera moves; start and end frames. | 21 |
| Grok Imagine 1.5 | xAI's Grok video: lively motion, with sound. | 29 |
 FLUX 3 | Crisp, stylish frames from Black Forest Labs, up to 20 seconds. | 35 |
| Model | Best for | Credits |
|---|---|---|
| Nano Banana Pro | Best at complex scenes and text inside images. Up to 4K. | 7 |
| Nano Banana 2 | Close to Pro quality, faster and cheaper. 1K or 2K. | 4 |
| Nano Banana | Quick everyday images and edits at a low price. | 2 |
| GPT Image 2 | Follows prompts precisely; clean typography. | 3 |
| Seedream 5.0 Pro | Sharp 2K images; strong at edits with several images. | 3 |
| Seedream 5.0 Lite | Cinematic, realistic light at a low price. 2K. | 2 |
| Seedream 5.0 Flash | The quickest Seedream: 2K drafts and fast edits. | 2 |
| Seedream 4.5 | Detailed 2K product shots and marketing images. | 2 |
| Seedream 4.0 | Steady 2K images and edits at a low price. | 2 |
| FLUX 3 | The newest FLUX: crisp detail and natural light. 1K or 2K. | 2 |
| FLUX Schnell | Fastest and cheapest. Good for drafts and ideas. | 1 |
| Ideogram 3.0 | Strong at posters, logos and words inside images. | 3 |
Image models: lowest credits per image. Video models: credits per clip at the model's default length and resolution, without optional sound. The composer shows the exact credits before you run.
How to Choose an AI Model
Start from what the result needs. These AI models overlap, but each one is clearly stronger at something.
Text inside the image
For posters, packaging and signs, GPT Image 2 and Ideogram 3.0 spell short text reliably. Nano Banana Pro handles longer text and infographics at up to 4K.
Video with sound
Veo 3.1 can add dialogue and sound effects, Seedance 2.0 adds sound at no extra credits, and Grok Imagine and MiniMax H3 Max always include sound.
Speed and credits
Nano Banana 2 is our default AI image model: about 20 seconds and 4 credits per image. For quick video drafts, Seedance 1.5 Pro makes a 5 second 720p clip for 6 credits.
Cinematic camera work
For big camera moves and action, compare Seedance 2.0 and Kling 3.0 on the same prompt. For one long take, Seedance 2.5 runs up to 30 seconds.
AI Models FAQ
Common questions about the models on ImageToPsd.AI.
Try the Top AI Models in One Place
Pick a model, write a prompt or add a photo, and see the credits before you run. Most images arrive within a minute and most clips within a few minutes.