AI Text to Image Generator

Yes, you can start for free. Sign up, then claim your first daily check-in: new accounts get a 10-credit welcome bonus on top of the regular check-in reward, so you have 12 free credits on day one, and 2 to 6 more free credits every following day. The lightest models, such as Z-Image Turbo and Qwen Image 2.1, cost 1 credit per image, so day one alone covers about a dozen images. Check-in credits expire after 3 days. For heavier use, premium models and 2K/4K output, see Pricing for credit packs and plans.

More than 40 text-to-image models, all in one place: Nano Banana 2 and Nano Banana Pro, GPT-Image 2.5, Seedream 5.0, FLUX.2 (klein, flex, pro and max), Qwen Image, Z-Image Turbo, Ideogram V3, Kling Image, Wan and others. Each model shows its credit cost before you generate, so you can compare speed, style and price. The Models page has details on every one.

It depends on the model and settings. Fast models start at 1 credit per image, while top-tier models and higher resolutions cost more. The exact cost is shown before you submit, and credits are only charged for images that are delivered; failed tasks are refunded automatically.

Paid users own the images they generate and can use them commercially, for marketing, products, social media and client work. Images made on the free tier fall under the free-tier terms in our Terms of Service. Paid plans also remove the watermark and let you keep generations private.

Most models offer common aspect ratios such as 1:1, 16:9, 9:16, 4:3 and 3:4, and several support custom sizes. Standard output is around 1K; models like Nano Banana 2, Nano Banana Pro and Seedream 5.0 can go to 2K or 4K, which is available on Pro plans.

Describe the subject first, then the setting, style, lighting and camera or art medium. For example: 'a golden retriever in a sunlit meadow, oil painting, warm evening light, visible brush strokes'. If you need readable text in the image, put the exact words in quotes and pick a model that handles typography well, such as GPT-Image 2.5, Ideogram V3 or Seedream 5.0.

Yes. Many models understand prompts in Chinese, Japanese, Korean, Spanish, French, German and other major languages, and the interface is available in 17 languages. For the most predictable results across every model, English prompts still work best.

Fast models such as Z-Image Turbo or FLUX.2 klein usually return images in a few seconds; high-end models and 4K output can take up to a minute or two. You can keep working while tasks run, and every result is saved to your history.