GPT-Image-2.6 Flare: Pricing, Complexity Control, SVG and GIF Output

What GPT-Image-2.6 Flare is

GPT-Image-2.6 Flare is an image generation and editing model from OpenAI, released on 4 November 2026 with the model id gpt-image-2.6-flare. OpenAI describes it as built for fast iteration and recommends it for interactive image generation and editing workflows, where a person tries several versions in a row. It takes text and images as input and returns images; audio and video are not supported. It is one of two GPT Image 2.6 models: GPT-Image-2.6 Sunburst has the same prices and features and is described for detailed visual work. On LMArena's image leaderboard GPT-Image-2.6 Flare has 1384 Elo (rank 4, medium effort).

Complexity, markup edits, SVG and GIF

OpenAI documents four features for the 2.6 models, shared by Flare and Sunburst. A complexity field sets how much the model plans before rendering: low, medium or high, with high used when the field is left out. OpenAI notes that lower complexity can suit focused edits such as changing a background or a sign, that higher complexity can help with dense text, complex layouts and relationships between objects, and that higher complexity can add latency and unwanted changes; it carries no extra fee. Markup-guided edits take an annotation layer marking where a change should happen, sent to the image edit endpoint, at the same token rates as other edits. SVG output is requested with output_format set to svg on the generations endpoint, one image at a time, and returns static vector content with no scripts or embedded raster images; the API renders a reference image first and bills both that image and the authoring tokens. Animated GIF output works on both the generations and edits endpoints, with frame size, frame count and number of animations limited by the model's pixel budget.

Pricing

OpenAI bills GPT-Image-2.6 Flare by tokens rather than per image. Text input costs $5 per million tokens, with cached text input at $1.25. Image input costs $8 per million tokens, with cached image input at $2. Image output costs $30 per million tokens. The number of tokens an image uses depends on its size and quality setting, so the price of a single picture varies with those choices; OpenAI's image generation guide lists the token counts.

Rate limits and safety identifiers

OpenAI does not publish fixed rate limits for GPT-Image-2.6 Flare on its model page. Limits follow the account usage tier, which rises automatically as an organisation sends more requests and spends more on the API, and each account can see its own limits in its settings. For apps that serve individual end users, OpenAI's image generation guide recommends sending a stable, privacy-preserving safety_identifier with each generation or edit request: a hash of a username or email address rather than the address itself, up to 128 characters. The older user parameter on the Images API is deprecated but still accepted, and a non-empty safety_identifier takes precedence over it.

How Flare fits the lineup

Because Flare and Sunburst cost the same per token, OpenAI's documentation separates them by the kind of work: Flare for quick, repeated rounds of generation and editing, Sunburst for detailed compositions. The previous generation, released on 8 September 2026, has its own Flare, which OpenAI describes as its fastest model for high-quality everyday image generation, at the same token prices. GPT-Image-2.5 Flare has 1398 Elo (rank 3) and GPT-Image-2.6 Sunburst has 1406 Elo (rank 2, medium effort). These are LMArena's figures; the site does not recommend one.

Official Resources