OmniLLM is a robust, production-grade Model Context Protocol (MCP) server designed to bridge Google Antigravity IDE and other MCP-compatible clients with powerful large language models such as Anthropic Claude, OpenAI GPT-4o, and Google Gemini. It offers dynamic multi-model routing, ensures persistent conversation memory via SQLite, and includes a sophisticated glassmorphic dashboard for live token streaming, cost tracking, and provider health monitoring, enhancing the development and management of AI applications.
Características Principales
01Auto-Router for dynamic best-model selection based on task
02Premium Glassmorphic Dashboard with live feeds and custom UI
03Real-time Token Streaming to dashboard and IDE
04Persistent multi-turn conversation memory via SQLite Context Chaining
05Per-request token counting and cost estimation
061 GitHub stars
