gpt-image-2.5-flareImageExamples verified 2026-09-25GPT Image 2.5 Flare
The everyday pick: about 19 seconds at 1024² on high, with xhigh / max tiers available.
GPT Image 2.5 Flare (gpt-image-2.5-flare) is the everyday GPT Image 2.5 model: about 19 seconds at 1024² on high, with xhigh and max tiers available, up to 10 images per request and up to 16 reference images. It is billed by tokens at an official price of $5 input and $30 output per 1M tokens; at 1024², low uses about 200 output tokens (about $0.006 per image) and high about 1,760 (about $0.0528). Call POST /v1/images/generations or /v1/images/edits.
- 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
- 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.
- Higher quality means more output tokens: at 1024², low measured about 200 and high about 1,760 output tokens.
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
- POST
/v1/images/generations - POST
/v1/images/edits - GET
/v1/images/tasks?task_id={task_id}
Quick start
Replace sk-your-key with a key whose group includes this model.
curl https://api.hop-base.com/v1/images/generations \
-H "Authorization: Bearer sk-your-key" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-image-2.5-flare",
"prompt": "A shiba inu under a cherry tree, Japanese watercolor",
"size": "1024x1024",
"quality": "medium",
"output_format": "png"
}' \
| jq -r '.data[0].b64_json' | base64 --decode > result.pngFAQ
- How much does one image cost?
- It is billed by tokens: $5 input, $1.25 cached input, and $30 output per 1M tokens. Output tokens grow with quality: at 1024², low uses about 200 (about $0.006 per image) and high about 1,760 (about $0.0528). Actual charge = official price × group multiplier.
- What size values are accepted?
- Only auto or WIDTHxHEIGHT; 1K / 2K / 4K return 400. The longest side is ≤ 3840, sides are multiples of 16, the aspect ratio is ≤ 3:1, and total pixels are 0.66–8.29 MP.
- Why did a failed request return 200?
- When synchronous generation takes longer than about 40 seconds, the server sends 200 first and keeps the connection alive, so a later failure is still a 200. Check whether the response body contains error, and set the client read timeout to at least 300 seconds.