Intelligent LLM routing across 200+ AI models for cost, quality, and latency optimization
OpenAI-compatible chat completions with Martian-specific router parameters
Anthropic-compatible Messages API support
Model listing with pricing and reliability information
MEAI AIFunction tools for integration with any IChatClient
All modern .NET features - nullability, trimming, NativeAOT, etc.
Usage
1 2 3 4 5 6 7 8 91011121314
usingMartian;usingvarclient=newMartianClient(apiKey);// Chat completion with cost optimizationvarresponse=awaitclient.CreateChatCompletionAsync(model:"openai/gpt-4.1-nano",messages:[newChatCompletionMessage{Role=ChatCompletionMessageRole.User,Content="Hello!",}],maxCost:0.01f,willingnessToPay:0.1f);
Chat Completion
Basic example showing how to create a chat completion via the Martian Gateway.
1 2 3 4 5 6 7 8 910
usingvarclient=newMartianClient(apiKey);// Send a chat completion request through the Martian Gatewayvarresponse=awaitclient.CreateChatCompletionAsync(model:"openai/gpt-4.1-nano",messages:[newChatCompletionMessage{Role=ChatCompletionMessageRole.User,Content="What is 2 + 2?",}]);
List Models
Example showing how to list all available models with pricing information.
1234567
usingvarclient=newMartianClient(apiKey);// List all available models on the Martian Gatewayvarresponse=awaitclient.ListModelsAsync();// Each model includes pricing and reliability informationvarfirstModel=response.Data[0];
Cost-Optimized Routing
Example showing how to use Martian's router parameters for cost-optimized model selection.
1 2 3 4 5 6 7 8 9101112131415
usingvarclient=newMartianClient(apiKey);// Use the router with cost constraints to optimize model selection.// The models parameter restricts which models the router can choose from.// willingness_to_pay controls cost vs. quality trade-off.varresponse=awaitclient.CreateChatCompletionAsync(model:"openai/gpt-4.1-nano",messages:[newChatCompletionMessage{Role=ChatCompletionMessageRole.User,Content="Explain the concept of machine learning in one sentence.",}],models:["openai/gpt-4.1-nano","openai/gpt-4.1-mini"],maxCost:0.01f,willingnessToPay:0.1f);