GPT Image 2.5 API
OpenAI's top image model — precise edits, perfect text, up to 4K
Try GPT Image 2.5
Runs on your balance at the price shown on the button. Results are saved to your account.
Output image shape.
Output size tier: 1K (≈1 MP), 2K (≈4 MP) or 4K (≈8 MP, experimental).
Rendering effort. Higher quality uses more output tokens and costs more; xhigh and max are for final assets.
Create a free account with Google or email — pay only for what you generate.
Your image will appear here
Write a prompt, pick your settings and hit Run. You only pay for successful generations.
What is GPT Image 2.5?
GPT Image 2.5 (Sunburst) is OpenAI's highest-quality image model: precise multi-image editing, reliable text and layout, five quality levels and output up to 4K.
Edits that stay on-brand
GPT Image 2.5 Sunburst prioritizes editing precision: change a background, a color or a detail and the product, label and faces stay exactly as they were. Combine up to 16 reference images.
Text you can ship
Posters, packaging, menus and UI mockups come out with correct spelling and clean layout, in any of 11 aspect ratios.
GPT Image 2.5 pricing
Pay per image. No subscription. Failed generations are refunded automatically.
| Tier | Official price | Cheaper Gen. | You save |
|---|---|---|---|
1K · low 1024×1024 | $0.0066/img | −11% | |
2K · low 2048×2048 | $0.012/img | −12% | |
1K · medium 1024×1024 | $0.013/img | −12% | |
2K · medium 2048×2048 | $0.025/img | −12% | |
1K · high 1024×1024 | $0.048/img | −12% | |
2K · high 2048×2048 | $0.096/img | −12% |
- Official price: OpenAI API list price for gpt-image-2.5-sunburst ($30 / 1M image output tokens, $5 / 1M text input, $8 / 1M image input). Prices verified 2026-10-16.
- Per-image figures are for square images and include a typical prompt; each input image adds about $0.016.
- xhigh and max quality cost more — see the price in the playground.
Specifications
- Resolutions
- 1K · 2K · 4K
- Quality
- low · medium · high · xhigh · max
- Aspect ratios
- 1:1 · 3:2 · 2:3 · 4:3 · 3:4 · 5:4 · 4:5 · 16:9 · 9:16 · 21:9 · 9:21
- Edit inputs
- Up to 16 images
- Model id (OpenAI)
- gpt-image-2.5-sunburst
- Released
- Sep 8, 2026
Use GPT Image 2.5 via API
One request to start a job, one to fetch the result. Same shape for every model.
import os, time, requests
API = "https://cheapergen.com/api/v1"
headers = {"Authorization": f"Bearer {os.environ['CHEAPERGEN_API_KEY']}"}
# 1. Submit the job
job = requests.post(f"{API}/run/openai/gpt-image-2.5/text-to-image", headers=headers, json={
"prompt": "A minimalist poster for a jazz night called “BLUE HOUR”, saxophone silhouette, deep blue and brass, clean typography",
"aspect_ratio": "1:1",
"resolution": "1K",
"quality": "medium"
}).json()
# 2. Poll until it finishes (or pass "webhook_url" and skip polling)
while job["status"] in ("queued", "running"):
time.sleep(3)
job = requests.get(f"{API}/generations/{job['id']}", headers=headers).json()
print(job["status"], job["output"]["images"][0]["url"] if job["status"] == "succeeded" else job["error"])Example prompts
Hit “Try it” to load a prompt into the playground above.
Packaging design for an oat-milk carton called “OATLY GOOD”, pastel illustration, nutrition panel with legible text
A UI mockup of a meditation app home screen, soft gradients, three cards with short labels
- Image Edit
Replace the background with a sunlit Mediterranean terrace; keep the bottle and label exactly the same
GPT Image 2.5 FAQ
Related models

GPT Image2.5 Flare
Text → Image · Image Edit
The faster GPT Image 2.5 for everyday images and volume

Nano BananaPro
Text → Image · Image Edit
Gemini 3 Pro Image — studio-grade generation and editing up to 4K

FLUX3
Text → Image · Image Edit
Black Forest Labs' newest image model — grounded, layout-aware, up to 4K
GPT Image 2.5 is developed by OpenAI. Cheaper Gen. provides API access through official upstream endpoints and does not own the model. Trademarks belong to their respective owners. · Markdown · OpenAPI
