OpenLuxGoogle models API
OpenLux carries the Gemini family: Gemini 3.1 Pro for multimodal reasoning, the 3.8 and 3 Flash tiers for high-throughput work, and the Pro and Flash image previews for generation — all 91-92% below official list rates.
Calling Google models on OpenLux
Gemini models combine a large context window with multimodal input and, in the Pro tier, explicit thinking modes for problems that need deliberation. The Flash tiers trade some depth for speed and price, which makes them the natural default for chat at scale, classification, extraction and long-document work.
The image side of the family is separate but equally useful: the Gemini 3 Pro image preview targets fidelity and mixed image/text tasks, while the 3.1 Flash image preview is the batch tier for generating many assets at $0.0122 per request.
Everything runs through the OpenLux endpoint with the OpenAI request shape, so a Gemini model is a model-name change away from whatever you are already calling. Rates are per token for text and per request for images, metered in USD with real-time cost in the console.
Why teams call Google on OpenLux
Multimodal understanding
Text, image and mixed input on the same model, which suits document, chart and screenshot workflows.
Thinking modes
Gemini 3.1 Pro supports deliberate reasoning for problems where a fast answer would be wrong.
Cheap throughput
Gemini 3.8 Flash at $0.0551 per 1M tokens and 3 Flash preview at $0.0368 make high-volume traffic affordable.
Image tiers
Pro-quality generation at $0.0243 and batch generation at $0.0122 per request, on the same key.
Rates for Google models
| Model | Category | Rate | Discount |
|---|---|---|---|
| gemini-3.1-pro-preview | Text | $0.1471 per 1M tokens | -92% |
| gemini-3.8-flash | Text | $0.0551 per 1M tokens | -92% |
| gemini-3-flash-preview | Text | $0.0368 per 1M tokens | -92% |
| gemini-3-pro-image-preview | Image | $0.0243 per request | -91% |
| gemini-3.1-flash-image-preview | Image | $0.0122 per request | -92% |
Rates are USD, metered pay as you go with no subscription. See the full model catalogue and pricing plans for credit allowances.
Where Google models fit
Document processing
Read long PDFs and mixed media, extract structured data and answer questions over the result.
High-volume chat
Serve assistants, classification and moderation at a rate that keeps unit economics positive.
Visual content
Generate campaign visuals and catalogue imagery with the image previews under the same account.
Google on OpenLux questions
How much do Gemini models cost on OpenLux?
Gemini 3.1 Pro is $0.1471 per 1M tokens, Gemini 3.8 Flash is $0.0551, Gemini 3 Flash preview is $0.0368 (all 92% below list), the 3 Pro image preview is $0.0243 and the 3.1 Flash image preview is $0.0122 per request (91-92% below list).
Do I need a Google Cloud project?
No. OpenLux serves the Gemini models through its own endpoint, so an OpenLux key is the only credential required.
Which Gemini model fits my workload?
Pro for reasoning and multimodal analysis, 3.8 Flash for everyday production traffic at scale, and the image previews when the output is a picture rather than text.
Is the request format different from OpenAI?
No. Gemini models on OpenLux accept the OpenAI chat-completions format, so existing SDKs and tools work without changes.
Start calling Google models today
One OpenLux key covers Google and every other provider in the catalogue, metered in USD with real-time cost in the console.