GPT Image 2.6 Flare API Docs
High-speed commercial text-to-image and 1–10 reference image editing (gpt-image-2.6-flare) across 15 aspect ratios, 1K/2K/4K resolutions, and 5 quality tiers.
Request schema
Supports all 225 output combinations (15 sizes × 3 resolutions × 5 qualities), transparent PNG cutouts (background: "transparent" with output_format: "png"), and up to 10 HTTPS reference images in image_urls. Unknown fields return HTTP 400 invalid_parameter.
| Field | Type & Default | Constraints & Description |
|---|---|---|
AuthorizationRequiredHeader | string | Bearer cm_live_... server-side API key from your account. |
Idempotency-KeyOptionalHeader | stringDefault: recommended | 1 to 255 ASCII chars. Prevents duplicate submissions and charges for 24h. |
modelRequired | "gpt-image-2.6-flare" | Stable public model ID for GPT Image 2.6 Flare. |
promptRequired | string | Non-empty UTF-8 prompt describing the desired image. |
sizeOptional | 15 aspect ratiosDefault: "1:1" | "1:1", "16:9", "9:16", "4:3", "3:4", "3:2", "2:3", "5:4", "4:5", "21:9", "9:21", "2:1", "1:2", "3:1", "1:3". |
resolutionOptional | "1K" | "2K" | "4K"Default: "1K" | Output resolution tier across all 15 ratios. |
qualityOptional | "low" | "medium" | "high" | "xhigh" | "max"Default: "low" | 5 rendering quality tiers; higher quality emits more output tokens. |
output_formatOptional | "png" | "jpeg" | "webp"Default: "png" | Output image format (use png or webp when background is transparent). |
backgroundOptional | "auto" | "transparent" | "opaque"Default: "auto" | Background alpha mode. |
image_urlsOptional | string[] (1..10)Default: omitted | Optional 1 to 10 HTTPS reference image URLs for image editing. |
Text-to-image & reference edit examples
Omit image_urls for pure text-to-image, or pass 1..10 public HTTPS image URLs (including previously generated files.callmodels.dev links) for multi-image editing.
curl -sS -X POST "https://api.callmodels.dev/v1/images/generations" \
-H "Authorization: Bearer $CALLMODELS_API_KEY" \
-H "Idempotency-Key: img_req_20261115_01" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-image-2.6-flare",
"prompt": "Editorial studio photograph of a matte ceramic espresso cup on warm limestone, soft morning sunlight",
"size": "16:9",
"resolution": "1K",
"quality": "low",
"output_format": "png"
}'Task response payloads
Poll GET /v1/tasks/:task_id every 2 seconds until status is completed or failed.
{
"id": "task_01e561ba-859e-47fe-be79-60bc6f1d6499",
"model": "gpt-image-2.6-flare",
"object": "image.generation.task",
"status": "pending",
"created_at": "2026-11-15T09:49:18.000Z"
}Token rates & standard 1K reference prices
Final billing settles strictly on actual token usage: image_output = 3.525 Credits / 1K tok ($0.03525), image_input = 0.94 Credits / 1K tok ($0.0094), and text_input = 0.5875 Credits / 1K tok ($0.005875). Standard 1K output estimates:
| 1K Spec (Ratio · Pixels · Quality) | Est. Credits | USD Equiv. | Official 1K |
|---|---|---|---|
| 16:9 · 1360 × 768 · low | 1.628550 Credits | $0.0163 | $0.040 |
| 9:16 · 768 × 1360 · low | 1.628550 Credits | $0.0163 | $0.040 |
| 1:1 · 1024 × 1024 · low | 2.904600 Credits | $0.0290 | $0.040 |
| 16:9 · 1360 × 768 · medium | 2.911650 Credits | $0.0291 | $0.070 |
| 9:16 · 768 × 1360 · medium | 2.911650 Credits | $0.0291 | $0.070 |
| 1:1 · 1024 × 1024 · medium | 5.107725 Credits | $0.0511 | $0.070 |
| 16:9 · 1360 × 768 · high | 3.401625 Credits | $0.0340 | $0.190 |
| 9:16 · 768 × 1360 · high | 3.401625 Credits | $0.0340 | $0.190 |
| 1:1 · 1024 × 1024 · high | 6.189900 Credits | $0.0619 | $0.190 |