
Gemini API
Gemini, Gemma, Imagen and Veo models from Google, through one OpenAI-compatible API.
Google DeepMind builds Gemini (multimodal models with very long context windows), the open Gemma models, Imagen for images and Veo for video. On UltraGPT you call them with the OpenAI SDK, no Google Cloud project required.
Models
13
From / 1M input
$0.25
Gemini 3.1 Flash Lite
Largest context
1.0M
Gemini 4 Argon
Google models and prices
Chat and embedding prices are USD per 1M tokens: input / output.
| Model | Type | Context | Price |
|---|---|---|---|
Gemini 4 Argongemini-4-argon | Chat | 1.0M | $2 / $10 |
Gemini 4 Carbongemini-4-carbon | Chat | 1.0M | $2 / $10 |
Gemini 4 Hydrogengemini-4-hydrogen | Chat | 1.0M | $1.5 / $7.5 |
Gemini 3.8 Flashgemini-3.8-flash | Chat | 1.0M | $0.75 / $3.75 |
Gemini 3.7 Flashgemini-3.7-flash | Chat | 1.0M | $0.75 / $3.75 |
Gemini 3.1 Progemini-3.1-pro | Chat | 1.0M | $2 / $12 |
Gemini 3.1 Flash Litegemini-3.1-flash-lite | Chat | 1.0M | $0.25 / $1.5 |
Nano Banana 2.1gemini-nano-banana-2.1 | Images | 66K | ~$0.040 / image |
Veo 3.1 Fastveo-3-1-fast | Video | — | $0.100 / sec |
Veo 3.1 Liteveo-3-1-lite | Video | — | $0.050 / sec |
Veo 3.1veo-3 | Video | — | $0.400 / sec |
Nano Banana 2 Litegemini-3.1-flash-lite-image | Images | 66K | ~$0.034 / image |
Gemini 3 Pro Imagegemini-3-pro-image-preview | Images | 66K | ~$0.137 / image |
Call Gemini 4 Argon in a few lines
Swap the model id for any model in the table above; nothing else changes.
curl https://api.ultragpt.pro/v1/chat/completions \
-H "Authorization: Bearer $ULTRAGPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-4-argon",
"messages": [{ "role": "user", "content": "Hello!" }]
}'Comparisons
FAQ
How do I get a Gemini API key?
Sign in to UltraGPT Developers, top up your balance and create an API key. The same key works for every Google model and every other model on UltraGPT - no separate Google account needed.
How much does the Gemini API cost on UltraGPT?
Chat models of Google start at $0.25 per 1M input tokens (Gemini 3.1 Flash Lite). You pay per token from a prepaid balance, with no subscription; each model's page lists its exact input, cached-input and output prices.
Does it work with the official SDK?
UltraGPT is OpenAI-compatible: use the official OpenAI SDK (Python, Node.js) or any OpenAI-compatible tool with the base URL https://api.ultragpt.pro/v1 and the model id from the table.
Which Google model has the largest context window?
Gemini 4 Argon (gemini-4-argon), with 1,048,576 tokens.
