LIVE
EU AI Act enforcement begins · June 2026NIST AI RMF — risk management framework publishedISO/IEC 42001 AI management standard now certifiableOpenAI o3 sets new reasoning benchmarksAnthropic raises $4B Series EEU AI Act enforcement begins · June 2026NIST AI RMF — risk management framework publishedISO/IEC 42001 AI management standard now certifiableOpenAI o3 sets new reasoning benchmarksAnthropic raises $4B Series EEU AI Act enforcement begins · June 2026NIST AI RMF — risk management framework publishedISO/IEC 42001 AI management standard now certifiableOpenAI o3 sets new reasoning benchmarksAnthropic raises $4B Series E

LiteLLM

Free
API Available

Open-source Python library to call 100+ LLMs with one unified OpenAI-format interface

BerriAI · Frameworks

Launched 2023

Visit Website

Quick Answer

LiteLLM is open-source python library to call 100+ llms with one unified openai-format interface, made by BerriAI. It is free and open source. Key uses: Unified LLM API abstraction, Self-hosted LLM gateway, Cost and budget management.

Last updated:

Users

25K+ GitHub stars; deployed at hundreds of enterprises

What is LiteLLM?

LiteLLM is an open-source Python library and proxy server that provides a unified interface to call 100+ large language models using the OpenAI format. It handles provider-specific parameter translation, streaming, token counting, budget management, fallbacks, and load balancing. LiteLLM Proxy runs as a standalone gateway that can be self-hosted, adding observability and cost controls to any LLM backend.

What's new in LiteLLM in 2026?

LiteLLM 1.x (2025): Virtual API keys, per-team budgets, spend tracking dashboard; 25K+ GitHub stars; LiteLLM Proxy widely deployed at enterprises; support for all Bedrock, Vertex AI, and Azure models; guardrails integration

How much does LiteLLM cost?

Open-Source

$0

  • Full Python library
  • LiteLLM Proxy server
  • All 100+ providers
  • MIT license
  • Community support

LiteLLM Enterprise

Custom

  • SSO/SAML
  • Audit logs
  • Priority support
  • SLA
  • Custom guardrails

What can you do with LiteLLM?

  • Unified LLM API abstraction
  • Self-hosted LLM gateway
  • Cost and budget management
  • Multi-provider fallbacks
  • Token usage tracking

Pros

  • 100+ providers unified
  • OpenAI SDK drop-in replacement
  • Self-hostable proxy
  • Budget limits and alerting
  • Detailed logging/observability
  • MIT license

Cons

  • Adds complexity for simple use cases
  • Proxy setup requires DevOps
  • Some model-specific features not abstracted

What are the best alternatives to LiteLLM?

Frequently Asked Questions about LiteLLM

How much does LiteLLM cost?

LiteLLM pricing: Open-Source at $0; LiteLLM Enterprise at Custom.

Is LiteLLM free?

Yes, LiteLLM is fully free and open source.

What is LiteLLM used for?

LiteLLM is used for: Unified LLM API abstraction, Self-hosted LLM gateway, Cost and budget management, Multi-provider fallbacks, Token usage tracking.

What are the best alternatives to LiteLLM?

Top alternatives to LiteLLM include LangChain, Amazon Bedrock, OpenRouter.

What platforms does LiteLLM support?

LiteLLM is available on: Python library, Docker (LiteLLM Proxy), Self-hosted server.

Details

Vendor

BerriAI

Category

Frameworks

Pricing Model

Free

Launched

2023

Website

litellm.ai

Underlying Models

Any (OpenAI, Anthropic, Google, Azure, Bedrock, Cohere, Hugging Face, Ollama, 100+ providers)

Platforms

Python libraryDocker (LiteLLM Proxy)Self-hosted server

Integrations

LangChainLlamaIndexOpenAI SDKAnthropic SDKLangfuseWeights & BiasesAll major LLM providers

Ideal For

  • ML engineers standardizing LLM access
  • Platform teams building AI gateways
  • Startups with multi-provider needs
  • Enterprise LLMOps

Tags

Open-SourcePythonLLM GatewayOpenAI-CompatibleProxyMulti-ModelSelf-Hosted