GPT Image 2
Save 70%Officialopenai/gpt-image-2
OpenAI GPT Image 2 — high-fidelity image generation & editing.
Try GPT Image 2 here
Tune the request, see the current price ceiling, and inspect the real response without leaving this model page.
- Input modalities
- TextImage
- Output modalities
- Image
- Capabilities
- Image generation · Image editing
- Upstream release
- 2026-04-21
- RouteMux name
- openai/gpt-image-2
What is GPT Image 2
GPT Image 2 is the image model OpenAI released on April 21, 2026 (snapshot gpt-image-2-2026-04-21), used for both generation and editing. On September 8, 2026, OpenAI released two GPT Image 2.5 models, Flare and Sunburst, and said Flare produces higher-quality images than GPT Image 2 at roughly half the latency.
On RouteMux, requests to the gpt-image-2 model name are served by GPT Image 2.5 Flare. The model name, endpoints and request shape stay the same, so existing code keeps working. It speaks the OpenAI Images protocol: /v1/images/generations and /v1/images/edits take the same request shape as OpenAI, and the image comes back as base64 in data[0].b64_json. Billing is per image, not per token: one image costs one fixed amount regardless of prompt length or scene complexity, so the budget is price times image count. The size parameter picks the aspect ratio rather than the resolution, and the dimensions in the response are the ones that count.
When to switch to a 2.5 model name: new integrations should call gpt-image-2.5-flare directly, and the xhigh and max quality settings and 2K / 4K output are only available under the 2.5 names. The gpt-image-2 name accepts quality up to high and rejects 2K / 4K sizes. When edit precision matters more than speed, use gpt-image-2.5-sunburst. The one reason to keep gpt-image-2 is existing code that has the name hard-coded.
GPT Image 2 API pricing
Prices verified 2026-06-08
Cache read $0.6/M · Cache write —/M
Estimate cost
Key capabilities
- Served by 2.5 Flare
- Requests to this model name are handled by GPT Image 2.5 Flare. Same API, no code changes.
- Flat price per image
- One price per image, independent of prompt length or scene complexity. Budget is price times count.
- Generation and editing
- Both
/v1/images/generationsand/v1/images/editsroute to this name, with the same request shape as OpenAI. - Quality up to high
- This name accepts auto, low, medium and high; xhigh, max and 2K / 4K need a 2.5 model name.
Limited time10% extra Credits on your first top-up — up to 200 bonus Credits
Applies to your first top-up within 72 hours of signing up. Minimum $20, once per account.
Interface and protocols
Use any of these IDs to call this model via the API.
openai/gpt-image-2Replace the ROUTEMUX_KEY placeholder with your API key. Create one →
from openai import OpenAI
client = OpenAI(
base_url="https://api.routemux.com/v1",
api_key="<ROUTEMUX_KEY>",
)
response = client.images.generate(
model="openai/gpt-image-2",
prompt="A red panda coding under the moonlight",
n=1,
)
print(response.data[0].url)Performance (last 30 days)
GPT Image 2 FAQ
›What is GPT Image 2?
OpenAI GPT Image 2 — high-fidelity image generation & editing.
›How much does the GPT Image 2 API cost?
Through RouteMux, GPT Image 2 is priced at $0.0186, below the official list price.
›How do I call GPT Image 2 via API?
GPT Image 2 is OpenAI-compatible: point your base URL at RouteMux and set the model field to openai/gpt-image-2 — no code changes needed.
›Which input and output modalities does GPT Image 2 support?
GPT Image 2 accepts text, image as input and produces image as output.
›When was GPT Image 2 released?
GPT Image 2 was released on April 21, 2026.
Sources
- OpenAI — GPT Image 2 model page ↗
- OpenAI — GPT Image 2.5 Flare model page ↗
- OpenAI — Image generation guide ↗
- OpenAI — API changelog ↗
Reviewed by the RouteMux team on 2026-10-24