akashtk18 

Switch AI providers from configuration, not code. ModelMux maps logical model profiles ("fast", "smart", "private") onto OpenAI, Google Gemini, Ollama, or any OpenAI-compatible endpoint, so application code... More information

Cost, token and latency tracking for ModelMux. Records what every LLM call cost, attributed per tenant and per feature, with verified pricing for Anthropic, OpenAI and Google Gemini models.