What is GPT Image 2?
GPT Image 2 is an image model from OpenAI. In practical terms, GPT Image is designed for prompt-led image generation and image editing across practical creative workflows. Text-to-image and reference image editing with 1K, 2K, or 4K output.
On I2V, GPT Image 2 supports Text to image, image to image. Available output options include 1K, 2K, 4K, with aspect-ratio behavior shown as auto, 1:1, 3:2, 2:3, 4:3, 3:4, 5:4, 4:5, 16:9, 9:16, 2:1, 1:2, 21:9, 9:21. The same public model ID and credit balance work in Studio and through the task API.
What GPT Image 2 can do
Input workflows
GPT Image 2 exposes Text to image, image to image. The model-locked Studio form only shows controls that apply to this model.
Output control
Configured output resolution: 1K, 2K, 4K. Supported aspect-ratio behavior: auto, 1:1, 3:2, 2:3, 4:3, 3:4, 5:4, 4:5, 16:9, 9:16, 2:1, 1:2, 21:9, 9:21.
Reference-led creation
Use up to 16 reference images where the selected mode accepts visual input.
Best uses for GPT Image 2
Campaign and social visuals
Turn a written concept into art direction, launch graphics, thumbnails, and social assets without changing tools.
Product and brand exploration
Test compositions, backgrounds, materials, lighting, and visual directions before committing to production.
Reference-guided iteration
Use existing images when the subject, layout, style, or product identity needs to carry into a new result.
When to choose GPT Image 2
Choose this model when you want one workflow for new image creation and directed edits, with resolution-specific pricing visible before generation. I2V shows the exact valid configurations before you run the task, so model selection can be based on both capability and cost.
GPT Image 2 FAQ
What is GPT Image 2?
GPT Image 2 is a OpenAI image model available in I2V Studio and through the I2V task API. It supports Text to image, image to image.
What inputs and outputs does GPT Image 2 support?
GPT Image 2 supports Text to image, image to image. Output resolution is 1K, 2K, 4K, with aspect-ratio behavior listed as auto, 1:1, 3:2, 2:3, 4:3, 3:4, 5:4, 4:5, 16:9, 9:16, 2:1, 1:2, 21:9, 9:21. The live form is restricted to valid options for this model.
How much does GPT Image 2 cost on I2V?
GPT Image 2 starts at 30 credits ($0.38 using the current Mini Pack reference) for Text to image, 1K. The pricing table lists other valid configurations.
Can I use GPT Image 2 through an API?
Yes. Use the public model ID gpt-image-2 with POST /api/v1/jobs/createTask, then poll GET /api/v1/jobs/recordInfo until the task succeeds or fails. Studio and API share one credit balance.
