GPT Image 2.5 is OpenAI’s image generation and editing family for product images, illustrations and campaign artwork. Choose Flare for fast creative exploration or Sunburst for detailed rendering and precise edits. On Vidgo, create images from text or edit with references and optional masks, with output up to 4K, adjustable quality and transparent backgrounds for your designs and applications.
GPT Image 2.5 is developed by OpenAI. Vidgo provides access to the model. Model names and trademarks belong to their respective owners.Choose Text to Image to create a new visual from a written prompt, or Image Edit to revise and combine existing images. Both workflows are available with Flare and Sunburst, so you can select a model around your generation speed and image precision requirements.
GPT Image 2.5 Flare Image Edit turns reference images and text prompts into new visuals, supporting 1–16 references and local repainting. It makes targeted changes quickly while preserving the original subject’s appearance and composition.
GPT Image 2.5 Flare Text to Image turns text prompts into detailed 1K–4K images, with transparent backgrounds and low latency. It combines fast responses with natural lighting while avoiding structural distortion in large images.
GPT Image 2.5 Sunburst Image Edit uses 1–16 references and masks for precise local edits and feature transfer across images. It locks subject identity and environmental lighting while changing selected elements, preserving quality throughout successive editing rounds.
GPT Image 2.5 Sunburst Text to Image turns complex prompts into visuals up to 4K, with fine material textures and realistic lighting. It accurately reproduces depth and fine detail while maintaining precise composition under long, detailed instructions.
Text to Image turns descriptions of subjects, scenes, styles and compositions into new images. Image Edit combines instructions, reference images and optional masks for scene composition, object replacement and local revisions. Start with your source material and intended result, then choose Flare or Sunburst and the output settings your finished image needs.
| Model / task | Input and workflow |
|---|---|
| GPT Image 2.5 Flare Image Edit | GPT Image 2.5 Flare Image Edit turns reference images and text prompts into new visuals, supporting 1–16 references and local repainting. It makes targeted changes quickly while preserving the original subject’s appearance and composition. |
| GPT Image 2.5 Flare Text to Image | GPT Image 2.5 Flare Text to Image turns text prompts into detailed 1K–4K images, with transparent backgrounds and low latency. It combines fast responses with natural lighting while avoiding structural distortion in large images. |
| GPT Image 2.5 Sunburst Image Edit | GPT Image 2.5 Sunburst Image Edit uses 1–16 references and masks for precise local edits and feature transfer across images. It locks subject identity and environmental lighting while changing selected elements, preserving quality throughout successive editing rounds. |
| GPT Image 2.5 Sunburst Text to Image | GPT Image 2.5 Sunburst Text to Image turns complex prompts into visuals up to 4K, with fine material textures and realistic lighting. It accurately reproduces depth and fine detail while maintaining precise composition under long, detailed instructions. |
GPT Image 2.5 combines text instructions, visual references and output controls throughout the creative process, from exploring compositions to refining details. Flare prioritizes generation speed, while Sunburst prioritizes image precision. Both models support image generation and editing.
Describe your subject, setting and visual style, then use Flare to generate compositions and creative variations. Its faster generation makes it suited to repeated prompt revisions, helping you establish a direction for social images, product concepts and illustrations before refining the selected idea.
Specify the direction of light, surface textures and spatial layers in your prompt, then use Sunburst for work that calls for greater precision. It suits product imagery, campaign artwork and complex scenes where people, objects and environments need close attention during detailed editing.
Upload references for people, objects or environments and explain the role of each image in your instructions. An editing task can use up to 16 reference images to guide subject appearance, scene composition and spatial relationships. Describe the features to preserve in subsequent revisions to keep each iteration focused on the same creative brief.
Upload the source image and a corresponding mask to identify the area you want to revise, then describe the intended change. The mask guides the editing region while the original image supplies the subject, lighting and composition, making this workflow useful for clothing adjustments, object replacement and scene details.
Choose low, medium, high, xhigh or max quality for the current stage of your work, and set the resolution to 1K, 2K or 4K. For transparent assets intended for interfaces, games or further compositing, select a transparent background and PNG or WebP format so the image is ready for the next production step.
These two recorded examples add a lantern to a courier’s hand in a snowy landscape and place a fern specimen on a botanist’s notebook. Reference image 1 supplies the person and scene, while reference image 2 supplies the object. The actual prompts describe placement, physical contact and details to preserve. Compare the inputs and results to see how subject appearance, object scale and lighting come together.
Use image 1 as the sole scene, person and composition reference. Preserve the courier's exact face, short curly hair, freckles, navy coat, orange satchel, body pose, snowfield and distant pine horizon. Use image 2 only as the design reference for the small old brass lantern. Place one matching lantern hanging naturally from the courier's left gloved hand at waist level, with her fingers wrapping around its handle and the lantern body below the hand. Adjust only the hand grip as necessary; retain the overall arm pose. The flame casts a restrained warm glow on the nearby glove, coat and snow, while the surrounding twilight remains cool violet. Keep the glass clear and brass worn, avoid enlarging the lantern. A quiet determined courier carrying light across the snow, one continuous scene, no new person or objects beyond the lantern. No brands, logos, products for sale, advertisements, promotional graphics, watermark, split screen or comparison panels.
Image 1 defines the botanist's exact identity, face, glasses, clothes, kneeling pose, basalt coast and warm misty afternoon lighting. Image 2 defines only the distinctive fern cutting and its frond structure. Place one matching fern cutting resting diagonally across the cream notebook in his right hand. His left index finger gently indicates the fern's small amber spore dots; keep the hands anatomically correct and preserve the overall arm positions. The fern should be physically supported by the notebook with fine contact shadows, naturally scaled to the notebook, and illuminated by the same warm coastal light. Keep the face, clothing, basalt rocks, ocean mist and scene composition unchanged. Do not copy the neutral background from image 2 or add other plants, people or writing. No brands, logos, products for sale, advertisements, promotional graphics, watermark, split screen or comparison panels.
Choose a workflow for product imagery, local retouching, transparent assets or design concepts. Supply visual references, describe the finished image and set quality and resolution for the current production stage, giving each generation a specific creative purpose.
Upload references for your product, props and setting, then describe their placement, scale and lighting to create scene images for ecommerce launches or campaign concepts. For a series of images, specify the product’s appearance and the brand details to preserve in each request.
Start with an existing image, mask the clothing, object or background area to adjust, and describe the new color, material or content. The original image provides composition and lighting context for portrait retouching, product variations and campaign revisions.
Generate characters, props and interface decorations from text, or isolate a subject from a reference image. Select a transparent background and save the result as PNG or WebP to prepare reusable assets for websites, game interfaces and image compositing.
Explore composition, color and style with different prompts and quality settings, then refine materials and lighting around the selected direction. Reserve higher quality settings for finished images that need close attention to detail, using the corresponding prices to plan your iteration budget.
Upload room, furniture and material references in one editing request, and describe the layout, object relationships and overall style. Up to 16 images can provide visual guidance for interior concepts, room arrangements and material combinations.
Choose an aspect ratio and 1K, 2K or 4K resolution for widescreen backgrounds, presentations and campaign artwork. For images with demanding detail requirements, combine these settings with xhigh or max quality to refine materials, lighting and depth.
GPT Image 2.5 is billed per output image, with the cost determined by the task, quality and resolution you choose. The table shows reference limits, maximum output resolution and starting prices for all four tasks. A starting price represents the lowest available configuration; select your intended quality and resolution on the task page to see the corresponding cost and plan your budget.
| Model | Provider | Workflow | Max reference images | Max output resolution | Starting price / image |
|---|---|---|---|---|---|
| GPT Image 2.5 Flare Image Edit | OpenAI | Image Edit | 16 | 4K | $0.00655 |
| GPT Image 2.5 Flare Text to Image | OpenAI | Text to Image | Not applicable | 4K | $0.00655 |
| GPT Image 2.5 Sunburst Image Edit | OpenAI | Image Edit | 16 | 4K | $0.00655 |
| GPT Image 2.5 Sunburst Text to Image | OpenAI | Text to Image | Not applicable | 4K | $0.00655 |
Try generation and editing on Vidgo, then follow the task documentation to bring the same creative brief into your application.
Enter your prompt on the task page and set the quality, resolution and background. For Image Edit, upload reference images and add a mask for local revisions. Run the task, review the subject, composition and details, then refine your prompt and settings to establish a suitable approach for your API integration.
Use the same GPT Image 2.5 task options from model selection and online testing through to API integration. Learn from real editing examples, check costs before submitting a request and follow the matching documentation to bring your tested creative brief into an application.
Review Flare and Sunburst generation and editing options together, then choose a model around your source material, creative goal and task specifications.
Open a task page, enter a prompt, upload references and configure the output. Review actual generation results before choosing the approach to use in your application.
Select quality and resolution on the task page to see the corresponding price per image and plan separate budgets for concept exploration and finished assets.
Compare references, actual prompts and outputs to understand how to describe subjects, object relationships and details to preserve when preparing your own request.
Use the returned task ID to check status and retrieve image URLs, or revisit online generations in your account history to review your creative process.
Follow the task’s documentation link for authentication instructions, parameter explanations and request examples to integrate model calls into your application.
GPT Image 2.5 is OpenAI’s image generation and editing family. It creates images from written descriptions and uses reference images for scene composition and local revisions. The family focuses on natural lighting, surface textures and preserving subjects during edits, with two model options: Flare for fast creative exploration and variations, and Sunburst for work that requires greater image precision and editing control. On Vidgo, you can try Text to Image or Image Edit online and use the corresponding API to bring generation into your application.
Read Vidgo integration guides and platform comparisons to learn about API usage, model selection and costs as you prepare your creative project or application.
