How to generate an image from a text prompt with AI
What is the best way to generate images from text with AI?
Pick an image model, write a prompt, set the size and run it. Nano Banana Pro, GPT Image and Seedream 5 Pro are the strongest general models. Ideogram 4.5 is the pick when the image needs legible text. Z-Image Turbo and Wan 2.6 Image cost the least for drafts and batches. In OTOY Studio you can run one prompt through several models side by side.
Models that do this
| Model | Maker | Pick it for | Takes | Credits per run |
|---|---|---|---|---|
| Nano Banana Pro | High-end, up to 4K | Prompt plus image | 14 to 28 (from €0.14) | |
| GPT Image 2.5 | OpenAI | Follows long prompts, renders text | Prompt plus image | 1 to 19 (from €0.01) |
| Seedream 5 Pro | ByteDance | ByteDance's flagship | Prompt plus image | 16 (from €0.16) |
| FLUX 2.0 | Black Forest Labs | Black Forest Labs' current model | Prompt plus image | 6 (from €0.06) |
| Ideogram 4.5 | Ideogram | Typography, posters and titles | Prompt plus image | 3 to 20 (from €0.03) |
| Qwen Image 3 | Alibaba | Alibaba's image model | Prompt plus image | 7 (from €0.07) |
| Z-Image Turbo | Alibaba | Very fast, very cheap drafts | Text prompt | 1 (from €0.01) |
The lowest-cost option here is GPT Image 2.5, from 1 credits (€0.01) per run. Credits are OTOY Studio's currency: 1 credit = €0.01. Prices are list prices before plan discounts (paid plans take 10% to 25% off every run). Every model is on every plan, including Free.
How to do it in OTOY Studio
- Add a text node with your prompt: subject, setting, style, light and framing.
- Add two or three image model nodes and connect the prompt.
- Set the size or aspect ratio and the number of images, then run.
- Keep the winner, then edit, upscale or animate it in the same workflow.
Related
Try it free: https://app.otoy.studio (the free plan includes 100 credits a month, no credit card).
Model prices are computed from OTOY Studio's billing configuration, exported 2026-10-20.