GPT Image 2

GPT Image 2 by OpenAI: $5 input and $30 output per 1M tokens, — context. Call it through the odnoga LLM gateway.

Vision

GPT Image 2 is served by OpenAI and called through odnoga with the same OpenAI-compatible request shape as every other model in the catalog. It accepts images alongside text. Cached input is billed at $1.25 per 1M tokens, so repeated prefixes cost less.

Specification and price

VendorOpenAI
Model IDgpt-image-2
Context window— tokens
Max output
CapabilitiesVision
Your plan
Per 1M tokensVendor costYour price on Free+7%
Input$5$5.38
Output$30$32.26
Cached input$1.25$1.34

Vendor cost is the list price per million tokens as recorded in the odnoga catalog; your price applies your plan margin with the same formula that bills every request — see pricing. Pricing

Call it through odnoga

const res = await fetch("https://api.odnoga.com/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.ODNOGA_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "gpt-image-2",
    messages: [{ role: "user", content: "Hello" }],
  }),
});

Same request shape for every vendor in the catalog — swap the model id and odnoga handles keys, routing, limits and cost accounting.

How to use GPT Image 2

Reviewed 2026-09-14

Best for

  • Product and marketing imagery generated from a text description, at flexible image sizes.
  • Editing an existing image while keeping the parts that must not change — the model takes high-fidelity image input.
  • In-product image features where users describe what they want in plain language.
  • Iterating on one visual detail at a time, which is how OpenAI recommends refining edits.

Not the right pick when

  • Text answers of any kind: this is an image model, not a chat model.
  • Pixel-exact reproduction of a logo or legal document — always have a human check the result.
  • Bulk generation without a cost ceiling; image calls are priced per image, not per token.

Practical tips through odnoga

  1. 01Describe subject, composition, style and constraints in that order, then refine one thing per iteration.
  2. 02For edits, say explicitly what must stay the same — that is what preserves detail across rounds.
  3. 03Set a per-tenant budget before exposing generation to end users; image spend climbs faster than text.
  4. 04Keep the prompt in a managed version so a change to house style is one version bump, not a code change.

What a month costs

1,000 calls a month, 10,000 input tokens and 2,000 output tokens each, at your Free price (vendor cost +7%):

Input (10M tokens)$53.76
Output (2M tokens)$64.52
Your cost per month on Free$118

Vendor list cost $110 + odnoga margin $8.28 (+7%). Cached input or answer reuse lowers it; odnoga records both numbers per request.

GPT Image 2 compared

ModelContextInput / 1MOutput / 1MCapabilities
GPT Image 2$5$30Vision
ChatGPT (chat-latest)400K$5$30Vision, Tools, JSON mode, Streaming
GPT Image 1 mini$2$8Vision

Questions

What can GPT Image 2 do?
OpenAI describes it as a state-of-the-art image generation model for fast, high-quality generation and editing, supporting flexible image sizes and high-fidelity image inputs.
How is it billed through odnoga?
Per generated image at the catalog rate, recorded in the same per-request ledger as text calls, so image and text spend appear together per tenant.
Can I keep a brand style consistent?
Put the style rules in a managed prompt version and reuse it for every generation; changing the style is then one version change with an audit trail.

All models · Pricing · Docs · Compare models in the evaluation lab

Sources: OpenAI — GPT Image 2 model page, OpenAI — image prompting guide

Where this model sits

One gateway, every model.