LiteLLM

LiteLLM serves as an LLM Gateway (OpenAI Proxy) that facilitates the management of authentication, load balancing, and expense tracking for more than 100 LLMs, all while adhering to the OpenAI format. This tool streamlines the utilization of LLM APIs from multiple providers, including OpenAI, Azure, Cohere, Anthropic, Replicate, and Google. LiteLLM ensures uniform outputs and handles exceptions across all LLM APIs, offering comprehensive logging and error tracking capabilities. Key features include cost monitoring, batch API processing, guardrails, model access management, budget tracking, LLM observability, rate limiting, prompt management, S3 logging, and pass-through endpoints.