GPT Image 2.5 Sunburst vs Flare: Which Model Should You Ship?

Coffee Summary

  • FACT (changelog Sep 8, 2026): OpenAI released GPT Image 2.5 Sunburst and GPT Image 2.5 Flare for Image API and Responses image generation.
  • Decision rule: Sunburst when edit precision matters; Flare for fast everyday generation.
  • Both support xhigh and max quality settings.
  • Both use GPT Image 2 token rates per the changelog — check live pricing.
  • Dated snapshots exist (…-2026-09-08) if you need pinning.

What happened

On September 8, 2026, OpenAI’s changelog announced two sibling image models:

  • gpt-image-2.5-sunburst — where editing precision matters most.
  • gpt-image-2.5-flare — for fast, high-quality everyday generation.

Both are available through the Image API and Responses API image generation tool, support xhigh and max, and are billed using GPT Image 2 token rates.

Why it matters

A single “best image model” slug is gone. Product teams now choose on latency vs edit fidelity, then tune quality. Pick Sunburst for demanding edits and Flare for high-volume everyday work.