Skip to content
gpt-image-2.6-sunburstImageExamples verified 2026-09-25

GPT Image 2.6 Sunburst

The quality-focused GPT Image 2.6 model: slower than Flare (about 66 seconds at 1024² on high), with the same SVG and animated GIF output.

Official price
$5 / $30/ 1M tokensInput / output
Size
Longest side ≤ 3840Sides in multiples of 16 · aspect ratio ≤ 3:1
Total pixels
0.66–8.29 MP
Images per request
1–10
Reference images
Up to 16Remote URLs ≤ 25 MiB each
Notes
  • size accepts only auto or WIDTHxHEIGHT; 1K / 2K / 4K returns 400.
  • Long synchronous jobs return 200 first and keep the connection alive, so failures also come back as 200; check the error field.
  • high and above take longer (about 90 seconds at 1024² on xhigh), so send Prefer: respond-async to run async; max at 1024² measured about 7,020 output tokens, about $0.21 per image (official price).

Pricing

Official price in USD. Actual charge = official price × group multiplier.

Input
$5 / 1M tokens
Cached input
$1.25 / 1M tokens
Output (incl. images)
$30 / 1M tokens

Specs and limits

Source: spec/openai.json · spec/official.json, the same data the gateway validates against.

Size
Longest side ≤ 3840Sides in multiples of 16 · aspect ratio ≤ 3:1
Total pixels
0.66–8.29 MP
Images per request
1–10
Reference images
Up to 16Remote URLs ≤ 25 MiB each
Prompt
≤ 32,000 characters

Endpoints

Same family

Prices come from the public catalog; limits come from specs exported from the gateway’s validation code. Actual charge = official price × group multiplier.