The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, and logging [Bedrock, Azure, OpenAI, Anthropic, OpenAI, VertexAI, vLLM, Nvidia NIM]
https://github.com/BerriAI/litellm-docs/blob/main/docs/guides/security_settings.md
Own shared HTTP mechanics, including case-insensitive header lookup, defaults, replacement, and transport
Do not choose provider credentials, OAuth policy, beta requirements, or model behavior
Apply provider-supplied header decisions without importing provider implementations
Preserve existing header precedence, duplicate handling, and unrelated forwarded headers
Verify observable request behavior rather than the structure of helper functions