Models
Image models
All image models in Clipia — developers, strengths, aspect ratios and resolution, how many photos they accept, modes and price in credits. How to choose one.
Clipia brings together image models from Google, OpenAI, ByteDance, Black Forest Labs, xAI, Midjourney and Alibaba. They differ in quality, speed, how well they render text, how many photos they accept and price. This page helps you compare them and pick the right one.
Every model works in Image Studio at /create-image. Many models also have their own page at /create-image/<model> with a description and examples.
Model comparison
| Model and developer | Credits per image | Best for |
|---|---|---|
| Nano Banana 2 · Google | 4–6 | Fast, accurate everyday images. Understands long descriptions and renders text |
| Nano Banana Pro · Google | 5–7 | Complex scenes, infographics and lettering. Keeps details and fonts precise |
| GPT Image 2 · OpenAI | 3–7 | Posters, lettering and edits by description. Neat text in the frame |
| GPT Image 2.5 Flare · OpenAI | 3–5 | Ad shots and posters with lettering. Carefully keeps photo details |
| GPT Image 2.5 Sunburst · OpenAI | 3–5 | Product shots and photorealism. Careful lighting and textures |
| GPT Image 1.5 · OpenAI | 6–24 | Mockups, interfaces and lettering. Crisp typography |
| Seedream 5.0 Lite · ByteDance | 3–4 | Affordable vivid images, up to 14 input photos |
| Seedream 5.0 Pro · ByteDance | 4–8 | Precise edits and infographics. Keeps the face and composition, handles Cyrillic lettering |
| Seedream 4.5 · ByteDance | 3 | Portraits and fashion, expressive lighting |
| Seedream V4 · ByteDance | 3 | Landscapes, architecture and illustrations |
| Grok Imagine · xAI | 2 | Fast creative drafts, bold ideas and styles |
| FLUX Kontext Pro · Black Forest Labs | 3 | Targeted photo edits: changes only what you asked |
| FLUX Kontext Max · Black Forest Labs | 4 | Complex edits and stylization, more precise in fine details |
| FLUX 2 Pro · Black Forest Labs | 3–4 | Photorealism and lifestyle, natural light and skin |
| FLUX 2 Flex · Black Forest Labs | 5–9 | Detailed illustrations, fine textures |
| Imagen 4 Fast · Google | 2 | Quick images at a low price, supports a negative prompt |
| Imagen 4 · Google | 4 | Food, objects and still life, natural lighting |
| Imagen 4 Ultra · Google | 5 | Maximum detail, close-up photorealism |
| Midjourney V7 · Midjourney | 8 | Artistic, atmospheric images with a recognizable style |
| Z-Image · Alibaba | 1 | Quick, affordable idea tests, vivid colors and fantasy |
| Qwen Image 2 · Alibaba | 6 | Images with text and single-photo editing |
| Wan 2.7 Image · Alibaba | 2–8 | Creating and combining product scenes |
The price range depends on the resolution or quality level. A breakdown by resolution is on the Image Studio page.
Prices
Prices on this page come from the service pricing and are refreshed together with the documentation. The exact cost is always shown on the launch button before you're charged.
Technical capabilities
| Model | Aspect ratios | Resolution or quality | Input photos | Image Studio modes |
|---|---|---|---|---|
| Nano Banana 2 | 14 ratios plus Auto, from 1:8 to 8:1 | 512px, 1K, 2K, 4K | up to 10 | text, edit, combine |
| Nano Banana Pro | 10 ratios plus Auto | 1K, 2K, 4K | up to 8 | text, edit, combine |
| GPT Image 2 | 8 ratios plus Auto, including 21:9 | 1K, 2K, 4K | up to 16 | text, edit, combine |
| GPT Image 2.5 Flare and Sunburst | 12 ratios plus Auto | 1K, 2K, 4K | up to 16 | text, edit, combine |
| GPT Image 1.5 | 1:1, 2:3, 3:2 | Medium, High | up to 16 | text, edit, combine |
| Seedream 5.0 Lite | 8 ratios, including 21:9 | Basic, High, Maximum (4K) | up to 14 | text, edit, combine |
| Seedream 5.0 Pro | 7 ratios | Basic, High (up to 2K) | up to 10 | text, edit, combine |
| Seedream 4.5 | 8 ratios | Basic, High | up to 14 | text, edit, combine |
| Seedream V4 | named sizes, from square to 21:9 | 1K, 2K, 4K | up to 10 | text, edit, combine |
| Grok Imagine | 1:1, 2:3, 3:2, 9:16, 16:9 | — | 1 when editing | text, edit |
| FLUX Kontext Pro and Max | 6 ratios or Match photo | — | 1 | text, edit |
| FLUX 2 Pro and Flex | 7 ratios or from the photo | 1K, 2K | up to 8 | text, edit, combine |
| Imagen 4 Fast, Imagen 4, Imagen 4 Ultra | 1:1, 16:9, 9:16, 4:3, 3:4 | — | — | text |
| Midjourney V7 | 11 ratios | — | — | text |
| Z-Image | 5 ratios | — | — | text |
| Qwen Image 2 | 5 ratios | — | 1 | text, edit |
| Wan 2.7 Image | 8 ratios, including 8:1 and 1:8 | 1K, 2K | up to 9 | text, edit, combine |
The Combine photos tab has an overall cap of 8 photos, even if a model accepts more.
Extra settings
- Render text in the frame: Nano Banana 2, Nano Banana Pro, GPT Image 2 and 2.5, Seedream 5.0 Pro, Qwen Image 2. Put the text in quotes.
- Negative prompt (what must not appear): Imagen 4 Fast and Imagen 4 Ultra.
- Seed for repeatable results: Imagen 4 Fast, Imagen 4, Imagen 4 Ultra, Qwen Image 2, Wan 2.7 Image.
- Description length: up to 20,000 characters for Nano Banana and GPT Image 2 and 2.5; up to 5,000 for Seedream 5.0 Pro, Seedream V4, Grok Imagine, FLUX Kontext, FLUX 2 Flex and Wan 2.7 Image; up to 4,000 for FLUX 2 Pro; up to 3,000 for GPT Image 1.5, Seedream 5.0 Lite and 4.5; up to 1,500 for Imagen 4 and Ultra; up to 1,024 for Midjourney V7; up to 1,000 for Imagen 4 Fast and Z-Image; up to 800 for Qwen Image 2.
How to choose a model
| Task | What to try |
|---|---|
| Your first image, everyday work | Nano Banana 2 — 5 welcome credits cover one image |
| Lettering, posters, banners, infographics | GPT Image 2, Nano Banana Pro, GPT Image 2.5 Flare. For Cyrillic lettering, Seedream 5.0 Pro |
| Photorealism and product shots | GPT Image 2.5 Sunburst, FLUX 2 Pro, Imagen 4 Ultra |
| Food and still life | Imagen 4 |
| Portraits and fashion | Seedream 4.5, FLUX 2 Pro |
| Artistic, atmospheric shots | Midjourney V7 |
| Illustrations with fine detail | FLUX 2 Flex, Seedream V4 |
| Targeted photo edits | FLUX Kontext Pro; for complex edits, FLUX Kontext Max or Seedream 5.0 Pro |
| A scene from many photos | GPT Image 2 (up to 16 photos), Seedream 4.5 and 5.0 Lite (up to 14) |
| 4K resolution | Nano Banana 2, Nano Banana Pro, GPT Image 2 and 2.5, Seedream V4, Seedream 5.0 Lite |
| Cheap idea exploration | Z-Image, Grok Imagine, Imagen 4 Fast, Wan 2.7 Image at 1K — the cheapest in the table above |
A practical approach: explore ideas on an inexpensive model, then make the final version on a stronger model at a higher resolution. In Image Studio, the Renders text, Up to 4K, Best quality and Fits your balance filters narrow the list right away.
Where else the models are available
- Image Studio (
/create-image) — every model on this page, with filters, price comparison and the Edit photo and Combine photos tabs. - Model pages at
/create-image/<model>— a description, examples and a shortcut to generating with that model. - Carousels — slide illustrations made with Nano Banana 2, on the Carousel tab of Image Studio.
Try it for free — 5 credits when you sign up.
How to choose a model
An overview of Clipia image, video and audio models — developers, how to read prices in credits and which model fits a task: quality, text, sound.
Video models
Passports of Clipia video models — Veo, Kling, Seedance, Wan, HappyHorse, Hailuo, Grok and more: modes, duration, quality, sound, prices and picks by task.