Skip to content
VYRA

Model families

GPT Image 2 online: OpenAI's image model with readable text

GPT Image 2 is OpenAI's image model. It is strongest where many image models are weak: words inside the picture (menus, packaging, posters, UI mock-ups) and following long, specific instructions. On VYRA you can use it from a browser on one credit balance, alongside Nano Banana 2.1, Seedream 4.5 and the video models.

A medium-quality image is 6 credits (about $0.12 on Pro); the slower high-quality setting is 22 credits. Upload up to 8 reference images to edit a photo or combine products and people; each reference adds about 2 credits (two references: 10 credits).

20 free credits on sign-up · failed jobs return credits automatically

Models that power it

ModelBest forFrom
GPT Image 2
OpenAI
OpenAI's sharpest images and typography, edits from up to 8 references8 cr
Medium · ≈$0.16
Nano Banana Pro
Google
Max reasoning, perfect typography, up to 4K15 cr
1K · ≈$0.30
Ideogram 3
Ideogram
Best for logos, posters & text6 cr
1 image · ≈$0.12
Nano Banana 2.1
Google
Google's newest image model, edits and up to 4K5 cr
1K · ≈$0.10

Prices in credits for the cheapest settings. ≈$ at the Pro rate ($7.99 for 400 credits). Studios show the exact price before you generate.

How to do it, step by step

  1. 1

    Open Image Studio with GPT Image 2

    The button on this page preselects GPT Image 2. Switch models at any time; each shows its price before you generate.

  2. 2

    Describe the image

    Put any words that must appear in quotation marks and say where they go: "headline at the top in bold sans-serif". Add reference photos for edits.

  3. 3

    Pick the aspect ratio

    1:1 for feeds and marketplaces, 9:16 for stories, 16:9 for banners and thumbnails.

  4. 4

    Generate, compare, download

    Make two or three variations, keep the best, and send it to Video Studio if you want it animated.

Example prompts

Starting points written for this task, not outputs we are showing you. Open one in the studio with its model selected, then adapt it.

  • Minimalist coffee shop menu board on a white wall, the words "MORNING MENU" at the top, three items with prices in a clean sans-serif, soft daylight, straight-on photo

    GPT Image 2 · from 8 crUse this prompt →
  • Flat lay of a skincare bottle on pale stone, label reading "HYDRA 24", morning shadows, editorial product photography, 4:3

    GPT Image 2 · from 8 crUse this prompt →
  • Mobile app onboarding screen for a habit tracker, headline "Small steps, every day", one illustration, green and cream palette

    GPT Image 2 · from 8 crUse this prompt →

What people make with it

Text in images

Posters, menus, labels and social graphics with spelled-out copy.

Photo edits

Change one thing in your photo from a written instruction.

Mock-ups

Packaging, app screens and signage concepts.

Product composites

Combine several product shots into one scene.

FAQ

Do I need a ChatGPT subscription?

No. VYRA calls GPT Image 2 through its own provider account; you pay VYRA credits per image (6 credits at medium quality, 22 at high).

Medium or high quality?

Medium is faster and covers most social and web use. High (22 credits) takes longer; use it for final assets with fine text or detail.

Can GPT Image 2 edit my own photo?

Yes. Add up to 8 reference images and describe the change. Each reference adds about 2 credits.

Which aspect ratios are available?

1:1, 9:16, 16:9, 3:4, 4:3.

Can I try it for free?

Yes. New accounts get 20 credits with no card, plus 20 free Flux 2 Flash image drafts a month. Whether that covers a job depends on the model and settings; the price is always shown first.

Can I use the results commercially?

Yes, on every plan, within the VYRA content policy (no real people without their consent, no impersonation, no deceptive deepfakes).

Related AI tools

Compare models

Try GPT Image 2 now.

Every model on one plan. Start with 20 free credits, upgrade when you need more.

Explore VYRA