LiteLLM

What is LiteLLM?

LiteLLM is an open-source gateway and Python SDK from BerriAI for working with many model providers through one interface. In code, the completion function takes the same arguments for OpenAI, Anthropic, Bedrock and 100+ others and returns results in the OpenAI format, with streaming, retries, fallbacks and a cost figure for every call. As a self-hosted AI Gateway it gives apps in any language a single endpoint, hands out virtual keys with budgets and rate limits, records each request and its cost, and keeps provider keys inside the gateway. The same deployment can expose MCP tools and A2A agents, balance load across deployments, apply guardrails and show an admin UI. An Enterprise license key adds SSO, SCIM, audit logs, team admins, secret managers, multi-region deployment and support SLAs. The software is free, and Enterprise pricing depends on deployment size.

Use cases

• Call many LLM providers through one API
• Issue virtual keys with budgets per team
• Track LLM spend by key, team and tag
• Route requests with fallbacks and load balancing
• Expose MCP tools from one endpoint
• Run Claude Code and Codex through a gateway
• Add SSO and audit logs to an AI platform

Who it's for

• Developers
• Platform engineering teams
• AI infrastructure teams
• Enterprises

Why use LiteLLM

• Switching models usually means changing one model string, which helps avoid provider lock-in.
• Keys, budgets and logs sit in one gateway, so spend stays visible per team and project.
• Provider credentials stay inside the gateway rather than in each application.
• The Enterprise tier is a license key on the same image, so there is nothing to migrate.
• Running in your own cloud keeps prompts and responses within your environment.

LiteLLM features

• OpenAI-format API for 100+ LLM providers
• Python SDK with streaming, retries and fallbacks
• Virtual keys with budgets and rate limits
• Spend tracking and cost per call
• Router with load balancing across deployments
• Guardrails and PII masking with Presidio
• MCP gateway and A2A agent calls
• Admin UI with pricing calculator
• SSO, SCIM and audit logs on Enterprise
• Secret manager support on Enterprise

LiteLLM integrations

• OpenAI
• Anthropic
• AWS Bedrock
• Langfuse
• Okta
• Google Workspace
• HashiCorp Vault
• MCP

How to get started with LiteLLM

1. Install the Python SDK, or run the gateway with the quickstart script or Docker image.
2. Add the providers and models you want to expose, with their API keys stored in the gateway.
3. Create virtual keys for teams or apps, with budgets, rate limits and allowed models.
4. Point your applications or coding agents at the gateway in OpenAI format.
5. Watch spend and logs in the admin UI, and add an Enterprise license for SSO and audit logs if needed.

LiteLLM limitations

• Self-hosted, so you run and secure the gateway yourself
• Budgets and virtual keys require a Postgres database
• Several guardrail integrations and SSO beyond 5 users need an Enterprise license
• Enterprise pricing is not published

LiteLLM pricing

LiteLLM is open source, so you can use and self-host it for free.

LiteLLM plans

• AI Gateway

  Free

  Open source, self-hosted

  • Virtual keys
  • Budgets and rate limits
  • Spend tracking
  • Admin UI
  • Load balancing
  • Guardrails
  • MCP gateway

• Enterprise

  Custom

  License key, priced by deployment size

  • SSO and SCIM
  • Audit logs
  • Org and team admins
  • Secret managers
  • IP allowlists
  • Multi-region
  • Support SLAs

Plans as listed on the official pricing page on 24 Oct 2026. Prices can change and may vary by country or exclude tax. Check the official pricing page.

LiteLLM alternatives

The main LiteLLM alternatives listed on Debuly are OpenRouter, Portkey, Helicone, Kong AI Gateway, Bifrost and OmniRoute. The table compares pricing and platforms; order is A–Z.

1. 

   Bifrost

   Open-source AI gateway from Maxim with one OpenAI-compatible API across providers, failover, budgets and an MCP gateway.

   Developer platformsAutomation and agents
   Open sourceVisit Bifrost (official site)
2. 

   Helicone

   Open-source LLM observability and AI gateway that logs requests, tracks cost and latency, and routes to 100+ models through one API.

   Developer platformsCoding
   FreemiumVisit Helicone (official site)
3. 

   Kong AI Gateway

   Gateway and control layer for LLM, MCP and agent traffic, with usage metering, billing and cost limits built in.

   Developer platformsAutomation and agents
   Contact salesVisit Kong AI Gateway (official site)
4. 

   OmniRoute

   Free open-source AI gateway that puts 352 providers behind one OpenAI-compatible endpoint with automatic fallback.

   Developer platformsCodingAutomation and agents
   Open sourceVisit OmniRoute (official site)
5. 

   OpenRouter

   Unified OpenAI-compatible API for 500+ AI models with provider fallbacks, pay-as-you-go credits and a free tier

   Developer platformsChat assistantsAutomation and agents
   FreemiumVisit OpenRouter (official site)
6. 

   Portkey

   AI gateway with observability, guardrails, caching and prompt management for teams running LLM apps in production.

   Developer platformsAutomation and agents
   FreemiumVisit Portkey (official site)

LiteLLM FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from docs.litellm.ai · fetched 24 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.