OpenAI released Images 2.6 on November 4, describing it as faster image generation, GIF creation, more precise editing and new API controls across ChatGPT, Codex and the API.
For developers, the release arrives as two new models: gpt-image-2.6-sunburst and gpt-image-2.6-flare. Both are served through the v1/images/generations and v1/images/edits endpoints. OpenAI's changelog lists adjustable complexity, SVG generation and animated GIF generation as the headline features, and adds markup-guided editing for supported GPT Image 2.5 and 2.6 models.
OpenAI's model documentation positions the two variants differently. Sunburst is aimed at detailed compositions and image editing workflows, while Flare is described as "image generation and editing for fast iteration," for interactive work where users revise images repeatedly.
Both models carry the same listed price per million tokens: $5 for text input ($1.25 cached), $8 for image input ($2 cached) and $30 for image output. Neither model is available on the Chat Completions, Responses or Batch endpoints, so applications must call the image endpoints directly.
For builders, SVG output is the more practical change: vector graphics can be edited, scaled and embedded in code without a separate tracing step. Because the two variants are priced the same, the choice between Sunburst and Flare comes down to whether a product needs detailed composition or fast iteration.