NousResearch Launches Hermes Agent: Open-Source Self-Improving AI With Cross-Platform Messaging and 200+ LLM Support
Summary
NousResearch launches Hermes Agent, an open-source, self-improving AI that learns from experience, supports 200+ LLMs, runs on infrastructure as cheap as a $5 VPS, and connects across Telegram, Discord, WhatsApp, Signal, and Slack — already earning 8.7k GitHub stars from 142 contributors.
Key Points
- NousResearch is releasing Hermes Agent, a self-improving AI agent featuring a built-in learning loop that creates and refines skills from experience, searches past conversations, and builds a persistent model of the user across sessions.
- Hermes Agent supports multi-platform messaging including Telegram, Discord, Slack, WhatsApp, and Signal, runs on flexible infrastructure from a $5 VPS to GPU clusters, and is compatible with over 200 LLM models via providers like OpenRouter, OpenAI, and Anthropic with no vendor lock-in.
- The open-source project, licensed under MIT and built by Nous Research, offers features including a cron scheduler, parallel subagent delegation, 40+ tools, MCP integration, and research-ready batch trajectory generation, with an active contributor base of 142 developers and 8.7k GitHub stars.