Free plan available

liteLLM

An open-source LLM gateway simplifying model access, spend tracking, and load balancing.

Visitlitellm.ai
Intro

What is liteLLM?

LiteLLM is an open-source LLM gateway and proxy designed to simplify how developers interact with over 100 large language models. By providing a unified OpenAI-compatible API format, it streamlines the process of calling various LLMs using a consistent `completion(model, messages)` syntax. Whether you are implementing advanced capabilities like function calling on gemini litellm, integrating with platforms like openrouter and google ai studio, or setting up a local litellm proxy alongside ollama, this tool handles authentication, load balancing, and spend tracking effortlessly. You can find its comprehensive source code on the litellm github repository and read full integration guides via the litellm docs.

liteLLM at a glance
Open Source is Free, contact for Enterprise pricing703K monthly visitsHas free access
Pricing

liteLLM Pricing Plans

Compare liteLLM free options, liteLLM paid pricing plans, and usage notes before you choose the best way to use this AI tool in 2026.

Open Source is Free, contact for Enterprise pricing

$0

Includes 100+ LLM provider integrations, Langfuse/Langsmith/OTEL logging, virtual keys, budgets, teams, load balancing, RPM/TPM limits, and LLM guardrails.

Contact for Pricing

For giving LLM access to a large number of developers and projects. Includes everything in OSS, plus Enterprise Support, custom SLAs, JWT Auth, SSO, and Audit Logs.

Pricing updated:Jun 11, 2026

Features

liteLLM AI Features

OpenAI-Compatible proxy and SDK for 100+ LLM providersAutomatic spend tracking, budgets, and rate limiting per key/user/team/orgLoad balancing and automatic LLM fallbacks for higher uptimeStandardized logging to external tools like Langfuse, Langsmith, and OpenTelemetryVirtual keys, JWT authentication, and SSO for secure team access
Pros & Cons

liteLLM Pros and Cons

Pros

  • Eliminates the need to transform inputs and outputs across different LLM providers
  • Saves significant engineering time when adopting newly released models
  • Robust logging and error tracking capabilities built-in
  • Highly scalable with support for docker deployments and Kubernetes

Limitations

  • Requires initial infrastructure setup and configuration for self-hosting
  • Advanced enterprise features require contacting the team for custom licensing

liteLLM FAQ

LiteLLM translates standard OpenAI format calls into the appropriate GCP/Gemini structure seamlessly. This makes complex tasks, like executing function calling on gemini litellm, incredibly straightforward without changing your core application logic.