Back to Models

GPT Image 2.5 Flare

openaiopenai/gpt-image-2.5-flare

OpenAI's image model tuned for fast, high-quality everyday image generation. Same per-token billing as GPT Image 2. Released September 8, 2026.

GPT Image 2.5 Flare is one of two image models OpenAI shipped alongside GPT Image 2's next generation, tuned for fast, high-quality everyday image generation. It shares GPT Image 2's per-token billing shape (metered text/image input, per-quality image output) and adds two vendor-published output-quality tiers, xhigh and max, beyond GPT Image 2's low/medium/high.

This model is not yet available to send traffic through TheRouter. It is catalogued here (matching TheRouter's openai/gpt-image-2.5-flare id) but ships is_active: false until a live call to OpenAI's /v1/images/generations and /v1/images/edits endpoints is verified from a production-capable environment β€” the sole hard precondition this change's own design imposes before routing any customer traffic to it. Only low/medium/high quality tiers are exposed through TheRouter's quality parameter today; xhigh/max are real, vendor-published prices but are not yet orderable through TheRouter until its billing schema is extended to bill them.

Best for
  • β€’ High-volume, everyday image generation where turnaround speed matters
  • β€’ Product mockups, marketing visuals, and rapid iteration on creative concepts
  • β€’ Once activated on TheRouter β€” the faster, lower-latency sibling to GPT Image 2.5 Sunburst for the same billing shape
Reach for something else if
  • β€’ Any live traffic today β€” the model is not yet activated on TheRouter (see overview)
  • β€’ Precision editing tasks requiring maximum fidelity β€” GPT Image 2.5 Sunburst is the sibling tuned for that
  • β€’ Requests that need the xhigh or max quality tier β€” not yet orderable through TheRouter's quality parameter
Context Length
--
Max Output
--
Image Priceper 1M tokens
$0.014/ image
Input Priceper 1M tokens
$5.40/ 1M tokens

Modalities

textimage→image

Capabilities

VisionImage GenerationImage Edit

Media Generation Capabilities

image_generation
sizes
  • 1024x1024
  • 1024x1536
  • 1536x1024
qualities
  • low
  • medium
  • high
max_outputs_per_request
4
output_formats
  • png
  • jpeg
  • webp
defaults
size
1024x1024
quality
medium
image_edit
sizes
  • 1024x1024
  • 1024x1536
  • 1536x1024
qualities
  • low
  • medium
  • high
max_reference_images
4
output_formats
  • png
  • jpeg
  • webp
defaults
size
1024x1024
quality
medium

Pricing Breakdown

TypeRate
Image$0.014 / image
Input$5.40 / 1M tokens
Output$32.40 / 1M tokens
Image input$8.64 / 1M tokens
Cached image input$2.16 / 1M tokens
Low quality$0.0065 / image
Medium quality$0.014 / image
High quality$0.0572 / image
X-high quality (not yet orderable via TheRouter's quality parameter)$0.1015 / image
Max quality (not yet orderable via TheRouter's quality parameter)$0.2279 / image

Price shown is medium-quality generation; low and high quality bill at their own per-image rate below. Two additional quality tiers this model supports upstream (xhigh, max) are not yet orderable through TheRouter's quality parameter.

Supported Parameters

promptsizequalitybackgroundoutput_formatnuser

Specifications

CapabilitiesText-to-image generation, image editingdevelopers.openai.com β†—verified
Output resolutions1024Γ—1024, 1024Γ—1536, 1536Γ—1024developers.openai.com β†—verified
Quality tiers (vendor-published)low, medium, high, xhigh, max β€” TheRouter currently exposes low/medium/high only (see overview)developers.openai.com β†—verified
Billing shapePer-token text/image input ($5/$8 per million tokens), per-quality image outputdevelopers.openai.com β†—verified

API Usage Examples

Use the global api.therouter.ai endpoint shown below for new integrations; the legacy China accelerated endpoint is retired.

Recommended: use the async API

Image generation typically takes 30–180s, beyond the edge sync timeout. The examples below use the ?async=true submit + poll pattern. Read the full async image generation & edit guide β†’

cURL
# 1) Submit job (returns 202 immediately with a polling URL).
# Image generation takes 30-180s β€” always use the async path in production.
JOB=$(curl -s -X POST "https://api.therouter.ai/v1/images/generations?async=true"   -H "Content-Type: application/json"   -H "Authorization: Bearer $THE_ROUTER_API_KEY"   -d '{
    "model": "openai/gpt-image-2.5-flare",
    "prompt": "A cinematic product render with soft studio lighting"
  }' | python3 -c "import sys,json;print(json.load(sys.stdin)['id'])")
echo "submitted: $JOB"

# 2) Poll until terminal (succeeded / failed / cancelled / expired).
while :; do
  R=$(curl -s "https://api.therouter.ai/v1/jobs/$JOB"     -H "Authorization: Bearer $THE_ROUTER_API_KEY")
  S=$(echo "$R" | python3 -c "import sys,json;print(json.load(sys.stdin)['status'])")
  echo "status: $S"
  case "$S" in
    succeeded) echo "$R" | python3 -c "import sys,json;print(json.load(sys.stdin)['unsigned_urls'][0])"; break ;;
    failed|cancelled|expired) echo "$R"; exit 1 ;;
  esac
  sleep 5
done

API guide

Image generation

GPT Image 2.5 Flare uses the Images API (not chat completions). Send a POST request to /v1/images/generations with your prompt and model ID. Not yet routable β€” see overview.

cURL
curl https://api.therouter.ai/v1/images/generations \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-image-2.5-flare",
    "prompt": "A flat-lay marketing photo of a running shoe on a bright orange background",
    "n": 1,
    "size": "1024x1024",
    "quality": "medium",
    "output_format": "png"
  }'

Image Editing Examples

Upload an image and describe the edit you want with a text prompt; the model returns the edited image as base64.

cURL
# Same async submit + poll pattern as /v1/images/generations.
JOB=$(curl -s -X POST "https://api.therouter.ai/v1/images/edits?async=true"   -H "Authorization: Bearer $THE_ROUTER_API_KEY"   -F "model=openai/gpt-image-2.5-flare"   -F "prompt=Turn this scene into a watercolor painting"   -F "size=1024x1024"   -F "image=@input.png" | python3 -c "import sys,json;print(json.load(sys.stdin)['id'])")
echo "submitted: $JOB"

while :; do
  R=$(curl -s "https://api.therouter.ai/v1/jobs/$JOB"     -H "Authorization: Bearer $THE_ROUTER_API_KEY")
  S=$(echo "$R" | python3 -c "import sys,json;print(json.load(sys.stdin)['status'])")
  echo "status: $S"
  case "$S" in
    succeeded) echo "$R" | python3 -c "import sys,json;print(json.load(sys.stdin)['unsigned_urls'][0])"; break ;;
    failed|cancelled|expired) echo "$R"; exit 1 ;;
  esac
  sleep 5
done

More from openai

Similar models

Cross-provider sibling models

Frequently asked

Can I send traffic to gpt-image-2.5-flare through TheRouter today?

Not yet. The model is catalogued but ships is_active: false pending a live-call verification of OpenAI's /v1/images/generations and /v1/images/edits endpoints. It will be activated in a follow-up commit once that verification succeeds.

How is Flare different from Sunburst?

Both share the same billing shape and vendor-published quality tiers. Sunburst is tuned for precision editing workflows; Flare is tuned for fast, high-quality everyday generation. Neither is yet activated on TheRouter (see overview).

Fact ledger β€” every claim on this page traces here
sourceURLretrieved
Capabilitiesdevelopers.openai.com β†—2026-09-17verified
Output resolutionsdevelopers.openai.com β†—2026-09-17verified
Quality tiers (vendor-published)developers.openai.com β†—2026-09-17verified
Billing shapedevelopers.openai.com β†—2026-09-17verified
How is Flare different from Sunburst?developers.openai.com β†—2026-09-17to verify
Customer Support