Describe the picture in your own words: what's in it, the style (photo, cartoon, painting…), the colors and the mood. More detail gets you closer to what you imagine.

Aspect Ratio

The shape of the result. 16:9 is wide, like YouTube or a TV. 9:16 is tall, like TikTok or phone stories. 1:1 is square, like an Instagram post.

P-Image 1
P-Image 2
P-Image 3
P-Image 4
P-Image 5
P-Image 6
P-Image 7
P-Image 8
P-Image 9

PrunaAI · Text to image

P-Image AI Image Generator

P-Image is PrunaAI's own text-to-image model, built to generate in about a second with close prompt adherence and good short text. Use P-Image to try many variations of an idea in a few minutes.

Real results

Discover What P-Image Can Create

Quote cards, pet photos, icons and ad concepts, each made from one prompt with P-Image. Unedited outputs from this page, with the prompt for each.

Photo of a golden retriever puppy sitting in autumn leaves, looking at the camera, soft afternoon light

Pet portrait · P-Image

Photo of a golden retriever puppy sitting in autumn leaves, looking at the camera, soft afternoon light

Key Features of P-Image

P-Image comes from PrunaAI, a company best known for its optimisation toolkit that makes other AI models run faster and cheaper. With P-Image, Pruna applied that expertise to a production model of its own, marketed for exact prompt adherence, good text rendering and a very low cost per image.

01 / 04

Generation in About a Second

Pruna describes P-Image as generating images in under a second. On ImageFX the wait is dominated by the network and queue rather than the model, so expect a few seconds end to end. That speed changes how you work: generate several variations, compare them, then refine the best prompt.

Try P-Image Now
Pet portrait
Prompt · Photo of a golden retriever puppy sitting in autumn leaves, looking at the camera, soft afternoon light

02 / 04

Close Prompt Adherence

P-Image is built to follow prompts closely, so drafts look like what you asked for. Write a clear, literal prompt and say exactly what should be in the frame and where. On ImageFX your prompt is sent as written.

Write a Literal Prompt for P-Image
Flat icon
Prompt · Flat icon of a cloud with a lightning bolt, rounded shapes, purple and white, centred, plain background

03 / 04

Short Text for Social Graphics

Short text renders better on P-Image than on most budget models, which makes it useful for quick quote images and social graphics. Put the words in quotes and keep them to a few.

Make a Quote Card with P-Image
Quote image
Prompt · Square image: "Coffee first" in a bold black sans-serif in the centre, a small steaming cup icon above it, pale yellow background

04 / 04

One of the Cheapest Models on ImageFX

P-Image is one of the lowest-priced image models on ImageFX, so you can explore an idea with many runs. Use P-Image for drafts and ad concepts, then rerun the winning prompt on a stronger model for the hero image.

Draft an Ad Concept with P-Image
Ad concept
Prompt · 16:9 ad concept: a red electric scooter on a city street at dawn, headline "Ride Quiet" in white at the top left

How to Use P-Image on ImageFX

  1. Write a Clear, Literal Prompt

    Open the P-Image generator at the top of this page. Say exactly what should be in the frame and where, and put any text in quotes.

  2. Pick the Shape and an Optional Seed

    Choose an aspect ratio. Under More Options you can fix a seed to repeat a result.

  3. Generate, Compare and Refine

    Click Generate, run a few variations, then download the PNG you like. Every image is saved to your account.

A P-Image run costs 2 credits; the Generate button shows the price first, and a failed run costs nothing. See plans and credit packs

FAQs About P-Image

What is P-Image?

P-Image (P-Image 1.0) is PrunaAI's own text-to-image model, built for very fast, low-cost production use. Its weights are closed; it is available through an API.

How fast is P-Image?

Pruna says under one second of generation time. That is the vendor's figure: the total wait on ImageFX includes upload and queueing, so expect a few seconds end to end.

Can P-Image render text?

Short text, yes; it is one of P-Image's advertised strengths. When text is the point of the image, Ideogram v3 Turbo is still the safer choice.

How does P-Image compare with Flux Schnell?

Both are low-cost drafting models. P-Image is newer and stronger on text and literal prompts; Flux Schnell has the FLUX look and open Apache 2.0 weights.

Does P-Image support LoRAs or prompt upsampling here?

No. P-Image supports custom LoRAs and an optional LLM prompt rewrite through its API, but ImageFX sends your prompt as written. There is no negative prompt either; describe what you want.

Is P-Image good enough for final images?

P-Image gives draft-level detail. For hero images, rerun the prompt on a stronger model such as FLUX.2 [max].

Can I fix the seed with P-Image?

Yes, under More Options.

Who makes P-Image?

Pruna (PrunaAI), a company that builds tools to make AI models faster and cheaper to run.

Try P-Image on ImageFX

Test ideas, prompts and layouts in seconds with PrunaAI's fast, low-cost image model.

Create with P-Image