Skip to main content
The Secure AI Gateway provides a unified interface to multiple top-tier AI models. You can specify the model using the model parameter in your API requests.
Model IDProviderDescription
openai/gpt-4o-miniOpenAIFast and cost-effective for simple tasks
openai/gpt-4oOpenAIMost capable OpenAI model
openai/gpt-5.2OpenAIThe best model for coding and agentic tasks
google/gemini-2.5-flashGoogleFastest and most cost-efficient multimodal model
google/gemini-2.5-proGoogleMost capable Google model with advanced reasoning
xai/grok-4-1-fast-reasoningxAI / ReasoningFast reasoning and problem solving
deepseek/deepseek-reasonerDeepSeekAdvanced reasoning model
deepseek/deepseek-chatDeepSeekFast and efficient chat model (V3)
openai/o1OpenAIAdvanced reasoning and problem-solving
minimax/MiniMax-M2.1MiniMaxReasoning model with tool support

​
Model Selection

To use a specific model, pass its ID in the request body: 
{
  "model": "deepseek/deepseek-reasoner",
  "messages": [...]
}

​
Capabilities Mapping

FeatureSupport
Text GenerationAll models
Visual Input (Vision)openai/gpt-4o, openai/gpt-5.2, google/gemini-2.5-pro, xai/grok-4-1-fast-reasoning
Web SearchAll models (via Gateway tool)
StreamingAll models

​
API Endpoints (OpenAI Compatible)

The Secure AI Gateway mimics the OpenAI API structure, meaning standard clients can fetch available models dynamically.

​
GET /api/v1/models

Retrieves a list of all currently supported models in the standard OpenAI response format. Example Request: 
curl http://localhost:3000/api/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY_HERE"

​
GET /api/v1/models/[model]

Retrieves information about a specific model. Model IDs with slashes (e.g. openai/gpt-4o) are fully supported. Example Request: 
curl http://localhost:3000/api/v1/models/openai/gpt-4o \
  -H "Authorization: Bearer YOUR_API_KEY_HERE"