Loading model details…
GPT Image 2.6 Flare generates and edits images from text and image inputs. Use it for interactive image generation and editing workflows.
OpenAI
GPT Image 2.6 Flare is OpenAI's fast-iteration image generation and editing model designed for interactive visual creation, frequent image generation, and flexible editing workflows. Its official API ID is gpt-image-2.6-flare. The model accepts text and image inputs and produces image outputs, supporting prompt-driven creation, reference-guided composition, visual transformation, localized modifications, and iterative refinement. It is designed for applications where responsive creative workflows, rapid experimentation, and continuous visual adjustments are important.
The model supports image generation and editing through the Image API, making it suitable for interactive design tools, AI image editors, marketing asset generators, and automated content production platforms. Its related image-generation workflows include complexity controls, markup-guided editing, SVG output, and animated GIF generation. These capabilities extend its applications from conventional image creation to vector graphics, targeted visual modifications, and animated content production.
| Feature | Benefit |
|---|---|
| Fast-Iteration Model Positioning | Designed for interactive image generation and editing workflows that involve frequent prompt adjustments and repeated visual refinement. |
| Multimodal Text and Image Inputs | Combines natural-language instructions with visual references for image generation, transformation, composition, and editing. |
| Dedicated Image API Endpoints | Supports v1/images/generations for image creation and v1/images/edits for modifying existing images. |
| Complexity Configuration | Provides additional generation controls for visual tasks with different levels of complexity. |
| Markup-Guided Editing | Uses visual annotations to communicate editing intent and guide changes to selected elements or regions. |
| SVG and Animated GIF Workflows | Extends conventional static image creation to scalable vector graphics and animated visual content. |
| Token-Based Pricing | Separately prices text input, image input, cached input, and generated image output for usage-based cost management. |
GPT Image 2.6 Flare is positioned for fast iteration and interactive image generation and editing, while GPT Image 2.6 Sunburst focuses on detailed visual work and precision-oriented editing. Flare is suitable for applications involving repeated prompt adjustments, rapid concept exploration, and continuous image refinement. Sunburst is better aligned with workflows prioritizing detailed visual instructions and complex editing requirements. Actual differences in latency, image quality, and editing precision should be evaluated through representative tests.
Yes. GPT Image 2.6 Flare accepts text and image inputs and supports image editing. Developers can provide reference images alongside natural-language instructions to describe desired changes to subjects, backgrounds, colors, composition, and visual styles. For complex compositions, prompts should clearly define the role of each reference and the intended relationships between elements.
The Complexity setting provides additional control over the complexity of image-generation tasks. Developers can use the supported configuration to adapt generation workflows to different creative requirements, from rapid visual exploration to more elaborate compositions. Exact parameter values and supported configurations should be verified against the official image-generation documentation.
OpenAI's image-generation documentation describes markup-guided editing workflows. Developers can use visual annotations to identify objects or regions that require modification and combine them with textual instructions describing the intended result. This approach is useful for localized editing, object replacement, and complex visual revisions, although markup should not be treated as a guarantee of pixel-perfect boundaries.
The official model page references image-generation documentation covering SVG and animated GIF output. SVG workflows are useful for scalable graphics and vector-based visual assets, while animated GIF workflows support short animations and dynamic content. Developers should confirm the supported output formats, animation parameters, and applicable restrictions for their chosen API integration.
OpenAI lists the following token-based API pricing:
| Pricing Category | Price per 1M Tokens |
|---|---|
| Text Input | $5.00 |
| Cached Text Input | $1.25 |
| Image Input | $8.00 |
| Cached Image Input | $2.00 |
| Image Output | $30.00 |
Actual request costs depend on token consumption and the selected image generation or editing workflow. These figures represent OpenAI's API reference pricing and may differ from pricing offered through LinkModel.
Open the live LinkModel model page and confirm its platform-specific model ID, generation and editing endpoints, reference-image input limits, Complexity settings, markup-guided editing parameters, SVG output, animated GIF generation, and pricing. Integrate against LinkModel's published request schema rather than assuming that OpenAI-native endpoints and parameters are directly interchangeable.
Output tokens are calculated from the quality grid, aspect ratio, and total pixel area. Estimated cost equals output tokens multiplied by the price per 1 million output tokens.
| Output specification | Resolution | Our pricing | Official |
|---|---|---|---|
| Low | 1024 × 1024 | $0.0160485 / image | $0.02469 / image |
| Low | 1024 × 1536 | $0.012948 / image | $0.01992 / image |
| Low | 1536 × 1024 | $0.012948 / image | $0.01992 / image |
| Medium | 1024 × 1024 | $0.0282555 / image | $0.04347 / image |
| Medium | 1024 × 1536 | $0.022074 / image | $0.03396 / image |
| Medium | 1536 × 1024 | $0.022074 / image | $0.03396 / image |
| High | 1024 × 1024 | $0.034242 / image | $0.05268 / image |
| High | 1024 × 1536 | $0.026754 / image | $0.04116 / image |
| High | 1536 × 1024 | $0.026754 / image | $0.04116 / image |
| XHigh | 1024 × 1024 | $0.060879 / image | $0.09366 / image |
| XHigh | 1024 × 1536 | $0.0479505 / image | $0.07377 / image |
| XHigh | 1536 × 1024 | $0.0479505 / image | $0.07377 / image |
| Max | 1024 × 1024 | $0.136968 / image | $0.21072 / image |
| Max | 1024 × 1536 | $0.107016 / image | $0.16464 / image |
| Max | 1536 × 1024 | $0.107016 / image | $0.16464 / image |