GPT Image 2.6 Flare
High-speed text-to-image and multi-image editing API across 15 aspect ratios × 1K/2K/4K resolutions × 5 quality tiers and up to 10 reference images, billed on actual tokens.
gpt-image-2.6-flarePOST /v1/images/generationsReference Prices & Token Rate Schedule
Every accepted request locks its published rate at submission. Below are standard 1K reference prices alongside an interactive 225-specification calculator (15 ratios × 3 resolutions × 5 qualities); final billing settles on actual token usage.
| Aspect ratio | Dimensions | Quality | Ref credits | USD equiv | Official price | Savings | Per $10 |
|---|---|---|---|---|---|---|---|
| 16:9 | 1360 × 768 | low | 1.628550 Credits | $0.0163 | -59% | ~614 images | |
| 9:16 | 768 × 1360 | low | 1.628550 Credits | $0.0163 | -59% | ~614 images | |
| 1:1 | 1024 × 1024 | low | 2.904600 Credits | $0.0290 | -27% | ~344 images | |
| 16:9 | 1360 × 768 | medium | 2.911650 Credits | $0.0291 | -58% | ~343 images | |
| 9:16 | 768 × 1360 | medium | 2.911650 Credits | $0.0291 | -58% | ~343 images | |
| 1:1 | 1024 × 1024 | medium | 5.107725 Credits | $0.0511 | -27% | ~195 images | |
| 16:9 | 1360 × 768 | high | 3.401625 Credits | $0.0340 | -82% | ~293 images | |
| 9:16 | 768 × 1360 | high | 3.401625 Credits | $0.0340 | -82% | ~293 images | |
| 1:1 | 1024 × 1024 | high | 6.189900 Credits | $0.0619 | -67% | ~161 images |
gpt-image-2.6-flareView all 15 published token ratesper 1,000 tokens
Settled charge = round(sum(actual_tokens * credits_per_1k / 1000), 6). Text-to-image calls meter text input and image output; passing image_urls also meters reference image input tokens.
| Metered item | Credits / 1K tok | USD / 1K tok | Billing scope |
|---|---|---|---|
| image_output | 3.525 Credits | $0.03525 | Generated PNG / JPEG / WebP output tokens (primary cost driver) |
| text_input | 0.5875 Credits | $0.00588 | Uncached prompt text input tokens |
| text_cached_input | 0.146875 Credits | $0.00147 | Cached prompt text input tokens |
| image_input | 0.94 Credits | $0.00940 | Reference image input tokens (when passing 1–10 image_urls) |
| image_cached_input | 0.235 Credits | $0.00235 | Cached reference image input tokens |
| svg_input_gpt_6_astra | 1.175 Credits | $0.01175 | Astra vector input metering (applied only when present) |
| svg_output_gpt_6_astra | 5.875 Credits | $0.05875 | Astra vector output metering |
| svg_input_gpt_6_sol | 0.235 Credits | $0.00235 | Sol vector input metering |
| svg_output_gpt_6_sol | 1.175 Credits | $0.01175 | Sol vector output metering |
| svg_input_gpt_6_luna | 0.01175 Credits | $0.00012 | Luna vector input metering |
| svg_output_gpt_6_luna | 0.05875 Credits | $0.00059 | Luna vector output metering |
| svg_input_gpt_6_1_sol | 0.235 Credits | $0.00235 | GPT 6.1 Sol vector input metering |
| svg_output_gpt_6_1_sol | 1.175 Credits | $0.01175 | GPT 6.1 Sol vector output metering |
| svg_input_gpt_6_1_terra | 0.1175 Credits | $0.00118 | GPT 6.1 Terra vector input metering |
| svg_output_gpt_6_1_terra | 0.5875 Credits | $0.00588 | GPT 6.1 Terra vector output metering |
Request Parameters & Code Examples
Send standard HTTPS POST requests to /v1/images/generations with your Bearer key and an Idempotency-Key, then poll /v1/tasks/:id for the 30-day result URL and exact credit charge.
| Parameter | Type / Default | Allowed values & notes |
|---|---|---|
AuthorizationRequired · header | string | Bearer cm_live_... server-side API key from your account. |
Idempotency-KeyOptional · header | stringDefault: recommended | 1 to 255 ASCII chars. Prevents duplicate submissions and charges for 24h. |
modelRequired · body | "gpt-image-2.6-flare" | Stable public model ID for GPT Image 2.6 Flare. |
promptRequired · body | string | Non-empty UTF-8 prompt describing the desired image. |
sizeOptional · body | 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 · body | "1K" | "2K" | "4K"Default: "1K" | Output resolution tier across all 15 ratios. |
qualityOptional · body | "low" | "medium" | "high" | "xhigh" | "max"Default: "low" | 5 rendering quality tiers; higher quality emits more output tokens. |
output_formatOptional · body | "png" | "jpeg" | "webp"Default: "png" | Output image format (use png or webp when background is transparent). |
backgroundOptional · body | "auto" | "transparent" | "opaque"Default: "auto" | Background alpha mode. |
image_urlsOptional · body | string[] (1..10)Default: omitted | Optional 1 to 10 HTTPS reference image URLs for image editing. |
# 1. Submit an idempotent task (HTTP 202 Accepted)
curl https://api.callmodels.dev/v1/images/generations \
-H "Authorization: Bearer $CALLMODELS_API_KEY" \
-H "Idempotency-Key: $REQUEST_ID" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-image-2.6-flare",
"prompt": "Stoneware espresso cup on warm travertine",
"size": "16:9",
"resolution": "1K",
"quality": "low"
}'
# 2. Poll status, 30-day signed URL, and settled credits
curl "https://api.callmodels.dev/v1/tasks/$TASK_ID" \
-H "Authorization: Bearer $CALLMODELS_API_KEY"HTTP/1.1 200 OK
X-Request-Id: req_94c8a29f8104
Cache-Control: no-store
{
"id": "task_01e561ba-859e-47fe-be79-60bc6f1d6499",
"object": "image.generation.task",
"model": "gpt-image-2.6-flare",
"status": "completed",
"created_at": "2026-11-15T09:49:18.000Z",
"usage": {
"cost_credits": "2.908713"
},
"results": [
{
"type": "image",
"url": "https://files.callmodels.dev/task_01e561ba.../0.png?expires=1797328183&sig=...",
"expires_at": "2026-12-15T09:49:43.000Z"
}
]
}Itemized Call Billing & 30-Day Result Delivery
Every credit hold, token count, and final charge is transparent and verifiable down to each API call in your console.
Sample call receipt · GPT Image 2.6 Flare
Verified Sample- Task ID
- task_01e561ba-859e-47fe-be79-60bc6f1d6499
- Submitted specification
- "Cat" · 1:1 · 1K · low · 1024 × 1024 PNG
- Initial credit hold
- 8.669150 Credits
- Text input
- 7 tok × 0.5875 / 1K = 0.004113 Credits
- Image output
- 824 tok × 3.525 / 1K = 2.904600 Credits
- Unused hold returned
- +5.760437 Credits
- Final settled charge
- 2.908713 Credits ($0.0291)
How billing and 30-day file delivery work
- 1Submitting a request locks the published token rate and temporarily holds the maximum credits for your chosen specification.
- 2Once generation completes, Callmodels verifies output file integrity before marking the task completed.
- 3Exact token usage is charged, any unused hold returns immediately to your balance, and a 30-day HTTPS link is signed.
- 4Your app or browser reads https://files.callmodels.dev/... directly for 30 days without passing an API key.
GPT Image 2.6 Flare FAQ
Why can the final credit charge differ slightly from the reference table?
GPT Image 2.6 Flare bills on actual input tokens and image output tokens. Reference tables and calculator estimates reflect standard output tokens; prompt text adds 0.5875 Credits / 1K uncached tokens and reference images add 0.94 Credits / 1K tokens.
Are 2K/4K resolutions, reference images, or batch fields like n=4 supported?
All 15 aspect ratios, 1K/2K/4K resolutions, 5 quality tiers (low to max), PNG/JPEG/WebP output, and 1–10 reference image_urls are natively supported (1 output image per request). Unknown fields such as n=4 return HTTP 400 invalid_parameter with 0 Credits held.
Am I charged if a request times out or fails?
Never. Whenever a task ends in failed, 100% of the held credits return automatically to your available account balance and API key quota.
Can I pass the returned files.callmodels.dev URL into a video generation model?
Yes. Result URLs include a 30-day signature and require no Authorization header, so you can pass them directly into image_urls for image-to-video tasks or render them in your UI.