LiteLLM
Unified gateway for 100+ LLMs with spend tracking
LiteLLM is a Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI format, with cost tracking, guardrails, loadbalancing and logging. Supports Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, VLLM, NVIDIA NIM.
Martian is a proprietary AI routing company that automatically forwards each LLM query to the model best suited for it, based on cost, performance, and reliability. It's a closed, commercial API layer — you don't get access to how the routing decisions are made or the ability to self-host the router.
LiteLLM is an open-source alternative for teams that want a unified gateway to 100+ LLMs with spend tracking, which they can self-host and inspect rather than depend on Martian's closed routing service.
FAQ
Is there a free, open-source alternative to Martian?
Yes. LiteLLM is open source and free to self-host as a unified LLM gateway.
What's the best open-source alternative to Martian?
LiteLLM is the closest open-source option for routing and managing calls across multiple LLM providers, though it doesn't include Martian's proprietary model-mapping routing algorithm.
What are the top open-source alternatives to Martian?
Browse all open-source alternatives to Martian above, or explore the full Developer Tools category for more options.