With LiteLLM Proxy Server, you'll get access to a Proxy Server to call 100+ LLMs in a unified interface where you'll be able to track spend, set budgets per virtual key and users. You'll be able to set budgets & rate limits per project, API key, and model on OpenAI Proxy Servers. You can also translate inputs to the provider's completion, embedding, and image_generation endpoints as well as retry/fallback logic across multiple LLM deployments.