What each model is
OpenAI describes GPT Image 2.5 Flare as its fastest model for high-quality, everyday image generation, and GPT Image 2.5 Sunburst as its most capable model for generation and editing, for workflows where editing precision matters most. Both accept text and image input, support the image generation and image edit endpoints including inpainting, and offer low, medium, high, xhigh, max and auto quality settings in the API.
Google describes Nano Banana Pro (model code gemini-3-pro-image) as a reasoning-driven engine for professional-grade image editing and generation. Google says it is best for complex graphic design, high-fidelity product mockups and factual data visualizations that need accurate text rendering and real-world grounding through Google Search. It takes text and image input, and Google prices its output at 1K/2K and 4K.
Side by side, from the official documentation
- Inputs: both take text and images.
- Editing: both edit existing images; OpenAI lists inpainting for GPT Image 2.5.
- Quality and size: OpenAI documents quality levels for GPT Image 2.5; Google documents 1K, 2K and 4K output for Nano Banana Pro.
- Search grounding: supported by Nano Banana Pro; not listed on the GPT Image 2.5 model pages.
- Batch processing: available for both, at lower rates than standard requests.
How the API list prices work
GPT Image 2.5 Flare and Sunburst share the same token rates: $5 per million text input tokens, $8 per million image input tokens and $30 per million image output tokens, or $15 with Batch. OpenAI notes that its GPT Image 2 calculator does not estimate GPT Image 2.5 token use, so there is no official per-image figure; the cost depends on the tokens a request consumes.
Google bills Nano Banana Pro image output at $120 per million tokens and converts that to about $0.134 per 1K or 2K image and $0.24 per 4K image. Text and image input cost $2 per million tokens. Batch output is listed at $0.067 per 1K/2K image and $0.12 per 4K image.
These are third-party API list prices for developers, checked on 2026-10-26. They are not what you pay in this studio.
What you can do in this studio
GPT Images runs GPT Image 2.5 Flare (selected by default) and Sunburst for text-to-image and for editing one uploaded image. Each run makes one PNG at medium quality. Flare costs 4 credits per image or 6 per edit; Sunburst costs 6 per image or 8 per edit. The studio does not silently switch to another model, and Nano Banana Pro is not available here.
Which one to choose
- Quick drafts and variations: GPT Image 2.5 Flare.
- A final pass or an edit where small details matter: GPT Image 2.5 Sunburst.
- Factual visuals that should draw on Google Search, or 4K output: Nano Banana Pro, according to Google's documentation.
- Undecided: run the same brief on Flare here and on Nano Banana Pro wherever you have access, then compare lettering and fine detail at full size.