OpenLuxImage generation & editing
Four image models, three providers, metered per request in USD. Generation, editing with reference images and batch pipelines all run through the same OpenLux key and the OpenAI images API shape.
Image models are priced per request rather than per token, which makes cost easy to reason about: a thousand product shots is a known multiple of the per-request rate. The four models below span roughly 3x in price, with the expensive tier buying higher fidelity and better prompt adherence rather than different endpoints.
Calls go to /v1/images/generations with the model name selecting the provider, so switching from a cheap iteration model to a premium final render is a one-line change. Editing-capable models additionally accept reference images, which covers retouching, style transfer and controlled variations.
The picks
4 models from the live catalogue, with the rate each one is metered at.
GPT Image 2 C
Lowest per-image rate, with editing and reference-image support — $0.00882 per request (-93% vs official rates).
Gemini 3.1 Flash Image Preview
Cheap batch generation from the Gemini flash line — $0.0122 per request (-92% vs official rates).
Gemini 3 Pro Image Preview
Highest-fidelity Gemini tier for mixed image and text tasks — $0.0243 per request (-91% vs official rates).
Image generation & editing questions
Which image models are available on OpenLux?
gpt-image-2-c at $0.00882 per request, Gemini 3.1 Flash image preview at $0.0122, FLUX 2 Pro at $0.021 and Gemini 3 Pro image preview at $0.0243.
Do these models support editing an existing image?
gpt-image-2-c and FLUX 2 Pro accept reference images for editing and controlled variations. The Gemini image tiers focus on generation from text prompts. Endpoint details live in the docs.
How are image requests billed?
Per request in USD, not per token. Each call appears in the console with its exact cost, so a batch of product renders can be priced before it runs.
How do I call an image model?
Through the OpenAI-compatible images endpoint on the same base URL and key as the chat models — the model field selects the provider. Existing OpenAI images code works after changing the base URL and key.
Compare the picks head-to-head
GPT Image 2 C vs Gemini 3 Pro Image Preview
Rates, strengths and which model to pick.
GPT Image 2 C vs Gemini 3.1 Flash Image Preview
Rates, strengths and which model to pick.
GPT Image 2 C vs FLUX 2 Pro
Rates, strengths and which model to pick.
Gemini 3 Pro Image Preview vs Gemini 3.1 Flash Image Preview
Rates, strengths and which model to pick.
Gemini 3.1 Flash Image Preview vs FLUX 2 Pro
Rates, strengths and which model to pick.
Gemini 3 Pro Image Preview vs FLUX 2 Pro
Rates, strengths and which model to pick.
Put these picks behind one key
OpenLux meters every model in this collection on the same OpenAI-compatible endpoint, with pay-as-you-go billing and free trial credit to start.