A hosted OpenAI-compatible and Google Developer API-compatible gateway for Google's Gemin i, Gemma, Imagen, OpenAI GPT, DeepSeek, and Anthropic Claude AI models.
To access the API, simply sign up on our website: 👉 Visit the Website
Every new account receives $5 in free credits upon registration!
Because our hosted API pricing is significantly cheaper than the official provider rate s (often up to several times cheaper), your $5 free credit has massive buying power —equivalent to a much larger amount of credits if spent directly with the official provid ers. You can run millions of tokens of high-quality queries entirely for free!
Instead of keeping static tables in this document, our active model lineup and pricing ar e updated live on our platform: 👉 Explore the Model Square
We support a wide array of state-of-the-art models, including:
- Google Gemini (including standard, grounding/search, lite, image, and TTS variants)
- Gemma (lightweight instruction-tuned models)
- Imagen (cutting-edge image generation)
- OpenAI GPT
- DeepSeek
- Anthropic Claude
👉 Join our Discord Server to stay updated, get supp ort, and connect with the community!
Our platform is engineered with a multi-protocol gateway, allowing you to interface with Gemini, Gemma, and other models in two distinct formats:
- OpenAI-Compatible API - Perfect for dropping into existing codebases, langchain, o r tools designed for OpenAI.
- Google Developer / Gemini API - Perfect for using official Google SDKs or calling native Gemini endpoints directly.
For complete API documentation, SDK integration guides, supported parameters, and code ex amples (Python, Node.js, cURL, etc.), please refer to our official platform documentation :
👉 View API Documentation & Usage Instructions (Check the Docs/API section on our platform)
All Gemini, Gemma, and Imagen models running on our platform support every capability t hey officially support out-of-the-box:
- 👁️ Multimodal & Vision – Process images, video, and audio payloads.
- 🛠️ Tool Calling & Advanced Integrations – Supports official tool structures, custom
function calling, and advanced built-in tools such as Google Maps (supported by
gemi ni-3.1-pro-previewand other capable models). - 🎨 Image Generation – High-fidelity generations via official Imagen models.
Google Search grounding can be enabled in two ways depending on the API format you use:
- Google Developer API (v1beta): Search grounding works natively using the standard official structure (passing the search tool parameter) on standard models without needing any special suffix.
- OpenAI-Compatible API: Since typical OpenAI clients cannot easily pass custom Goog
le Search parameters, we host dedicated
:searchsuffix models (such asgemini-3.5- flash:search). Simply querying a model with the:searchsuffix activates Google Search grounding automatically through standard OpenAI SDKs without any extra configuration.
For issues, questions, or updates, join our Discord Server.