GPT Image 2.5 Flare API
GPT Image 2.5 Flare is OpenAI's speed-first API model: GPT Image 2-level quality at roughly 50% lower latency, positioned by OpenAI as the default choice for most applications. Roko currently exposes generation and reference-image editing at $0.02 per 1K image, $0.03 per 2K, and $0.04 per 4K.
- Public model ID
gpt-image-2.5-flare- Output tiers
1K / 2K / 4K (1K default)- Billing
Per output by resolution- Endpoint
POST /v1/images/tasks- Catalog snapshot
What is GPT Image 2.5 Flare?
GPT Image 2.5 Flare is one of the two API image models OpenAI shipped with the ChatGPT Images 2.5 launch. Officially it is the “small model, optimized for speed”: quality comparable to GPT Image 2 at roughly 50% lower latency, and OpenAI calls it the default choice for most applications. Early customers praise that it “understands what not to change” — making meaningful edits while preserving character, composition, and visual identity.
OpenAI supports generation, editing, and transparent backgrounds upstream. Roko's current public form exposes generation and reference-image editing but no background or output_format control, so this page does not promise alpha-channel output.
Roko currently exposes 1K, 2K, and 4K tiers plus 14 size choices including auto at $0.02 / $0.03 / $0.04 per image. Calls run as async tasks — submit, then poll for the result.
Core capabilities
The official speed tier
GPT Image 2-level quality at roughly 50% lower latency — OpenAI's default image model for most applications.
Precise editing, subject preservation
OpenAI and early customers highlight that it knows what not to change: edit what matters while keeping character and composition.
Upstream transparency
OpenAI supports transparent backgrounds; Roko does not currently expose the dedicated controls, so request validation remains authoritative.
Built for volume
Low latency plus $0.02 starting pricing fits social content, visual search, and high-concurrency generation.
GPT Image 2.5 Flare API pricing
Pricing is generated from the current Roko catalog and billed per output at the chosen resolution. Confirm the console estimate before submitting.
| Resolution | Price | Billing unit |
|---|---|---|
| 1k | $0.02 | per output |
| 2k | $0.03 | per output |
| 4k | $0.04 | per output |
Cost examples
Standard cost for everyday content and fast iteration.
Typical spend for social/marketing batches; failed images are not billed.
Parameters and supported scope
These fields come from the current public generation_config. quality exposes auto only; capabilities outside the form follow request validation.
| Parameter | Values | Notes |
|---|---|---|
prompt | textarea | Describe the image to create or the changes to apply to references. · Required |
image_urls | url / base64 | Reference images may use URL or Base64. State what must stay and what may change. · Optional |
size | auto / 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 / 1:3 / 3:1 | Selects an aspect ratio, resolution tier, or explicit pixel dimensions as configured for the model. · Default: auto · Optional |
resolution | 1K / 2K / 4K | Selects an output resolution tier configured for this model; the value may directly affect price. · Default: 1K · Optional |
quality | auto | The quality tier follows the model config (auto / low / medium / standard / hd) and can directly affect price. · Default: auto · Optional |
Call GPT Image 2.5 Flare with Roko API
Run the async task flow: submit a task to get a task_id, then poll for the result. Never expose an API key in client-side code.
Submit generation
curl https://api.rokoapi.com/v1/images/tasks \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: YOUR_UNIQUE_REQUEST_ID" \
-d '{
"model": "gpt-image-2.5-flare",
"prompt": "Fast social campaign visual, bold readable headline, consistent brand colors, clean composition",
"size": "16:9",
"resolution": "2K",
"quality": "auto"
}'Query recovery task
curl https://api.rokoapi.com/v1/tasks/$TASK_ID \
-H "Authorization: Bearer $API_KEY"POST /v1/images/tasks returns 202 with a task_id; once generation completes, query GET /v1/tasks/{task_id} for the result or failure reason. Idempotency-Key must be unique; replays with the same key return the original task without re-billing.
Best-fit scenarios
Social and creator content
Low latency makes per-item interactive assets a viable workflow.
Vertical story graphic announcing a weekend sale, bold typography, warm gradient background.
Fast reference-based variants
Upload a reference, state what must stay, and use the speed tier for social and marketing variants.
Keep the coffee cup unchanged and replace only the background with a warm cafe counter.
Product experiences and visual search
An official target scenario: generation inside products, where response speed drives retention.
Generate a hero image matching this search query: minimalist desk setup with warm lighting.
High-volume marketing assets
From $0.02 per 1K image for A/B copy art and multi-variant campaigns.
Six ad creative variations for a plant delivery app, same layout, different hero plants.
Important limitations
- Quality matches — not exceeds — GPT Image 2; for brand finals and repeated refinement use 2.5 Sunburst.
- quality currently exposes auto only; capabilities outside the form follow request validation.
- A submitted task is not a successful generation — trust the task query result.
- OpenAI applies provenance measures (C2PA metadata, invisible watermarking) to its image products; output rights depend on your inputs and applicable terms.
Flare, Sunburst, or GPT Image 2?
Everyday and volume work picks Flare (faster, cheaper); brand finals and repeated refinement pick Sunburst (above GPT Image 2 quality); validated existing workflows can stay on GPT Image 2 ($0.016 per 1K is cheaper).
All models →GPT Image 2.5 Flare FAQ
What model ID should I use for GPT Image 2.5 Flare?
Use the public Roko API model ID gpt-image-2.5-flare exactly as shown in the model field.
How much does the GPT Image 2.5 Flare API cost?
The current catalog lists $0.02 per 1K output, $0.03 per 2K, and $0.04 per 4K. Prices can change; confirm the console estimate and final settlement.
Flare vs Sunburst — what's the difference?
Per OpenAI: Flare is the small model optimized for speed with quality comparable to GPT Image 2; Sunburst is the base model optimized for quality, above GPT Image 2. Both share generation, editing, and transparent backgrounds — the trade-off is speed versus fidelity.
How much faster is Flare than GPT Image 2?
OpenAI states roughly 50% lower latency, and early customers report two to four times the speed of GPT Image 2. Real numbers depend on prompts, references, and output size — benchmark on your own workload.
Does it support transparent backgrounds?
OpenAI supports it upstream, but Roko's current public form has no background or output_format field. This page therefore does not promise alpha-channel output; follow request validation and the actual response.
Which aspect ratios are supported?
The current form has 14 choices: auto, 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 5:4, 4:5, 21:9, 9:21, 2:1, and 1:2. The default is auto; request validation is authoritative.
Which endpoint does it use?
Submit async tasks via POST /v1/images/tasks to get a task_id, then poll GET /v1/tasks/{task_id} for the result or failure reason.
What does migrating from GPT Image 2 involve?
Swap the model field to gpt-image-2.5-flare — the form is nearly identical. OpenAI's suggested path: run the same prompts on Flare first, compare quality, then switch fully and measure the latency win.