Railway
Instant Deployments, Effortless Scale
Start new thread
Jonathan Aucancela

LLM Observatory - Real-time observability for Claude & OpenAI APIs

by•
LLM Observatory — Monitor your AI API usage in real-time

Add a comment

Replies

Best
Jonathan Aucancela
Hunter
📌
Drop-in SDK wrapper (Node.js + Python) that tracks tokens, cost, and latency for every Claude/OpenAI call — with zero latency overhead. Open-source, self-hostable, multi-tenant. WebSocket dashboard, budget alerts, Discord notifications, and data export included.