Build a new picture from a clear brief, or upload a photo and explain an edit. GPT Image 2 gives you a place to try scenes, layouts, and details with text or reference images. Choose the frame and output size, then review the credits before you generate. Start with one useful change and judge the draft at Everygen AI.
JPEG, PNG or WebP · Up to 10 images · 10 MB each · 30 MB total
Write the change you want in plain words. These tasks show ways to use GPT Image 2 with a subject, a setting, and a clear plan. Name what should stay, add the details that matter, and review the result before using it in a larger project.
Use GPT Image 2 for a new image or a photo edit. Start with the subject and the part of the scene that matters most. Add the setting, frame, and light, then review one draft before making a longer list of changes.
Create a small glass sculpture on a pale blue shelf. Give it a soft pink tint and a round base. Use studio light and a clean background.
Separate the main scene from details about color and layout. Explain where each object should go and which part should draw the eye. Read the result against that list to find the next useful edit.
Show a quiet bookstore with a red armchair near a large window, books on wooden shelves, and a sleeping cat. Keep the light soft and the aisle open.
For a poster, ask for a clear headline and a small set of supporting words. Specify spelling, placement, and contrast. Check the letters in the saved output before placing the artwork into a wider campaign.
Create a modern tea poster with a green cup and pale paper texture. Put SLOW MORNING at the top and a small leaf near the bottom.
Name the light source and the surfaces in the scene. A broad window creates a different mood from a hard flash. Include a clear focus point, natural shadows, and the amount of background blur you want.
Photograph a ceramic bowl of ripe pears on a wooden table beside a window. Use soft side light, gentle shadows, and natural colors.
Reference images from Raphael. The briefs here are our own.
Compare other tools for your next image. Each card opens its model page, where you can review examples, output options, and the current credit estimate.
The GPT Image 2 AI image generator uses the same workflow for a new scene and an edit. Set the goal, check the options, then review one result. A small, clear change makes the next draft easier to judge.
Describe the subject, setting, and change you want. Add source photos when they help show what to preserve. Use short, direct instructions and keep each file within the limit shown by the tool.
Choose an available model and its output size. Set the aspect ratio and file format. The total reflects the selected options and any added reference charge; review it before starting the request.
Open the output at full size and check the details that matter. Download the file when it fits your goal. To refine a draft, use it as the source for another edit with a focused new instruction.
Use your existing account credits. The workbench shows the estimate for the chosen model, size, and reference images. Each request creates one output; plan credits also cover other supported tools.
Pricing
Make a few images, bring a scene to life, or create every day. Start with a credit pack, then subscribe when you need more.
Pay once. Use your credits for images or videos whenever you need them.
10 credits
$9.90
≈ 5 videos
50 credits
$29.00
≈ 25 videos
Or about 3–25 images
350 credits
$99.00
≈ 175 videos
Or about 25–175 images
900 credits
$199.00
≈ 450 videos
Or about 64–450 images
3000 credits
$499.00
≈ 1500 videos
Or about 214–1500 images
Images and videos share one balance. Monthly and yearly plan credits are both released each month.
34 credits / month
$8.33mo
≈ 17 videos/month
Or about 2–17 images
Billed annually at $100.00
334 credits / month
$41.67mo
≈ 167 videos/month
Or about 23–167 images
Billed annually at $500.00
500 credits / month
$50.00mo
≈ 250 videos/month
Or about 35–250 images
Billed annually at $600.00
Payments are handled by our payment providers. We do not store full card numbers.
Prices are listed in USD. Confirm the payment method, currency, and final amount at checkout.
Credits per image. Choose a model and resolution; the generator shows the cost before you submit.
| Model | 1K | 2K | 4K |
|---|---|---|---|
| GPT Image 2 | 6 | 10 | 14 |
| Gemini 3 Pro Image | 8 | 8 | 14 |
| Gemini 3.1 Flash Image | 5 | 8 | 12 |
Credits per clip at the duration shown, from the same rate rules as the generator. Duration changes are prorated and rounded up. A dash means the resolution is unsupported; available settings and modes are shown in the generator.
| Model | Duration | 480P | 720P | 1080P | 2K | 4K |
|---|---|---|---|---|---|---|
| Veo 3.1 | 8s | — | 80 | 100 | — | 200 |
| Veo 3.1 Lite | 8s | — | 30 | 46 | — | — |
| Veo 3.1 Fast | 8s | — | 40 | 50 | — | 150 |
| Veo 3 | 8s | — | 70 | 100 | — | — |
| Veo 3 Fast | 8s | — | 30 | 40 | — | — |
| Seedance 2.5 | 5s | 20 | 40 | 60 | — | 200 |
| Seedance 2.5 Turbo | 5s | — | 25 | 30 | — | — |
| Seedance 2.0 | 5s | 15 | 30 | 75 | — | 150 |
| Seedance 2.0 Mini | 5s | 8 | 15 | 35 | — | 70 |
| Seedance 2.0 Mini Turbo | 5s | — | 10 | 12 | — | — |
| Seedance 2.0 Fast | 5s | 12 | 25 | 60 | — | 120 |
| Seedance 1.5 Pro Fast | 5s | — | 3 | 4 | — | — |
| Kling 3.0 | 5s | — | 15 | 20 | — | 50 |
| Kling 2.6 | 5s | — | 10 | 15 | — | — |
| Wan 3.0 | 5s | 5 | 15 | 20 | — | — |
| Wan 2.7 | 5s | — | 10 | 15 | 30 | 50 |
| Wan 2.2 Ultra Fast | 5s | 2 | — | — | — | — |
| Grok Imagine 1.5 | 6s | 12 | 20 | 30 | — | — |
| Grok Imagine | 6s | 10 | 10 | — | — | — |
Answers about GPT Image 2 prompts, reference files, output options, and the credit estimate at Everygen AI.
GPT Image 2 creates an image from text and supports edits guided by pictures. Describe the scene or the change, choose the options shown in the tool, and generate one result. The maker is OpenAI; the workbench displays the models currently enabled for your account.
Start with the main subject and its setting. Add the light, layout, and the exact words that should appear in the image. For edits, name both the change and the details to keep. A focused brief is easier to review than several unrelated goals.
Use JPEG, PNG, or WebP. You can add up to ten pictures, with a limit of 10 MB for each file and 30 MB in total. The source images are added to the request when you choose Generate. Pick files you have permission to use.
GPT Image 2 offers 1K, 2K, and 4K in this workbench. Choose the size, frame, and format in the settings panel. The tool shows a credit estimate for that choice. Actual pixel dimensions vary with the aspect ratio.
Check the estimate on the Generate button after choosing the model, size, and files. Some models charge for added references. A reference example only fills the draft; it starts a paid request when you choose Generate. Each request makes one output.
You can read the guide and try the draft controls before signing in. Generation needs an account, enough credits, and an enabled model. Your task result and its download use the same account that created the request.
Bring a clear idea to the GPT Image 2 AI image generator. Use a short brief or a source picture, check the frame and credit estimate, and make your next draft.
Start creating
