GPT Image 1.5
Detailed prompt generation guided by up to sixteen references.
- Workflows
- Text to Image · Image Edit
- Output
- Model managed
- References
- Up to 16
- Access
- Paid access required
GPT IMAGE / PROMPTS + REFERENCES
Create and edit with GPT Image models
Use GPT Image 1.5 or GPT Image 2 for prompt-to-image generation and reference-guided editing, with version-specific quality, aspect-ratio, and resolution controls.
Content verification
Method: Model and workflow specifications are drawn from the current RedVideo configuration catalog. Live availability, supported settings, and generation quotes can change; confirm them in Studio before generating.
MODELS AND CAPABILITIES
Specifications below come from the workflows currently configured in RedVideo. Live availability and generation quotes remain visible in the studio.
GPT Image 2
Latest GPT image model with nine ratio modes and up to 4K output.
- Workflows
- Text to Image · Image Edit
- Output
- 1K · 2K · 4K
- References
- Up to 16
- Access
- Paid access required
Choose quality controls or a broader output range
GPT Image 1.5 supports square, landscape, and portrait ratios with Medium or High quality choices. GPT Image 2 expands the ratio menu and offers 1K, 2K, and 4K resolution settings where compatible.
Both routes support text-to-image and image-edit workflows. Each accepts up to sixteen supported references, giving a request more visual context without guaranteeing that every reference detail will be preserved.
- GPT Image 1.5: Medium or High quality selection
- GPT Image 2: nine ratio modes and up to 4K settings
- Up to sixteen supported references on either route
Combine visual context with a clear edit instruction
Reference images can establish subjects, objects, composition, or visual direction. State what should change and what should remain visually important instead of relying on the references alone.
Generated typography, identity, fine geometry, and small details can vary. Review outputs before using them in a production asset or as a source frame for video.
Let the selected ratio shape the available resolution
GPT Image 2 resolves incompatible combinations in the studio. Auto ratio uses a 1K setting, and a square request does not retain the 4K choice in the current compatibility rules.
Check the live task quote after changing version, mode, references, quality, ratio, resolution, or output count. The displayed controls and quote are the current source for the request.
FREQUENTLY ASKED QUESTIONS
- Which GPT Image models are available in RedVideo?
- The current catalog lists GPT Image 1.5 and GPT Image 2.
- Can GPT Image edit an uploaded image?
- Yes. Both GPT Image 1.5 and GPT Image 2 support image editing with supported reference inputs.
- How many reference images can GPT Image use?
- Each listed GPT Image route accepts up to sixteen supported references in the current catalog.
- What output controls does GPT Image 1.5 provide?
- GPT Image 1.5 offers three aspect ratios and Medium or High quality choices.
- Does GPT Image 2 support a 4K setting?
- GPT Image 2 lists 1K, 2K, and 4K settings, subject to ratio compatibility. The studio normalizes combinations that the route does not accept.
OPEN REDVIDEO
BUILD THE FIRST GENERATION
Select the model and workflow that fit the idea, review the live settings and quote, then start creating.