| Website | litellm.ai |
| Category | AI / LLM Infrastructure |
| License | MIT (open source) |
| Pricing | Free (open source), Enterprise available |
Overview
LiteLLM is an open-source LLM proxy that provides a unified API for 100+ LLM providers. Load balancing, fallbacks, spend tracking. One API for all models.
Pros
- MIT license
- 100+ model provider support
- Unified OpenAI-compatible API
- Load balancing and fallbacks
- Spend tracking per team/key
- Self-hosted option
Cons
- Adds a proxy layer (latency)
- Configuration can be complex
- Enterprise features limited
- Rapid changes can break things
- Documentation sometimes lags
Verdict
LiteLLM solves the multi-model problem: one API that works with OpenAI, Anthropic, Google, Azure, AWS, and 100+ other providers. The load balancing and fallback features mean your app stays up even when a provider goes down, and the spend tracking gives you per-team cost visibility. For companies using multiple LLM providers, LiteLLM is essential infrastructure.