Mantis, A self-hosted LLM gateway
Another LLM gateway when LiteLLM and Portkey already dominate the space.
Self-hosted LLM gateway. Cost effective, deterministic, and fast. Secure and private by default.
Cache-aware LLM routing that doesn't burn prompts to save pennies.
Engineering teams using multiple LLM providers, cost-conscious AI deployments
LiteLLM · Portkey · Helicone
Another LLM gateway when LiteLLM and Portkey already dominate the space.
Semantic caching for LLM APIs exists (Anthropic prompt caching, Langchain, Miniplex, vLLM); gateway routing is table stakes.
Cache-aware LLM eval with self-hosted model support beats Ragas on flexibility.
Drop-in OpenAI API gateway with failover—LiteLLM does this but this has a dashboard.
Go gateway with circuit breakers, but auth isn't production-ready yet.
Sub-15ms failover and atomic budget caps, but LiteLLM and Portkey exist.