Hermes
Agent
MCP
Telegram
Discord
Slack
Streaming
Plugins
Profiles
Memory
Approvals
Voice
Docker
OAuth
Skills
Provider
Gateway
OPEN-SOURCE AI AGENT

Hermes Agent
Release Notes

NousResearch/hermes-agent

An open-source AI agent that runs across CLI, Telegram, Discord, Slack, WhatsApp, Signal, and 10+ more platforms. Track every release here.

v0.8.0
Latest
7
Releases
956+
PRs Merged
4 days ago
Last Release
▼ SCROLL
Hermes AgentMulti-PlatformMCP ClientStreamingPluginsProfilesMemoryVoice ModeDockerOAuth 2.1Skills Hub400+ ModelsApache 2.0Hermes AgentMulti-PlatformMCP ClientStreamingPluginsProfilesMemoryVoice ModeDockerOAuth 2.1Skills Hub400+ ModelsApache 2.0Hermes AgentMulti-PlatformMCP ClientStreamingPluginsProfilesMemoryVoice ModeDockerOAuth 2.1Skills Hub400+ ModelsApache 2.0

Release History

LATEST
v0.8.0Apr 8, 2026209 PRs / 82 issues
The intelligence release — background task auto-notifications, free MiMo v2 Pro on Nous Portal, live model switching across all platforms, self-optimized GPT/Codex guidance, native Google AI Studio, smart inactivity timeouts, approval buttons, MCP OAuth 2.1, and 209 merged PRs with 82 resolved issues.
  • Background Process Auto-Notifications (`notify_on_complete`)
  • Free Xiaomi MiMo v2 Pro on Nous Portal
  • Live Model Switching (`/model` Command)
  • Self-Optimized GPT/Codex Tool-Use Guidance
  • Google AI Studio (Gemini) Native Provider
GitHub
V0.7.0
v0.7.0Apr 3, 2026168 PRs / 46 issues
The resilience release — pluggable memory providers, credential pool rotation, Camofox anti-detection browser, inline diff previews, gateway hardening across race conditions and approval routing, and deep security fixes across 168 PRs and 46 resolved issues.
GitHub
V0.6.0
v0.6.0Mar 30, 202695 PRs / 16 issues
The multi-instance release — Profiles for running isolated agent instances, MCP server mode, Docker container, fallback provider chains, two new messaging platforms (Feishu/Lark and WeCom), Telegram webhook mode, Slack multi-workspace OAuth, 95 PRs and 16 resolved issues in 2 days.
GitHub
V0.5.0
v0.5.0Mar 28, 2026157 PRs / 0 issues
The hardening release — Hugging Face provider, /model command overhaul, Telegram Private Chat Topics, native Modal SDK, plugin lifecycle hooks, tool-use enforcement for GPT models, Nix flake, 50+ security and reliability fixes, and a comprehensive supply chain audit.
GitHub
V0.4.0
v0.4.0Mar 24, 2026280 PRs / 0 issues
The platform expansion release — OpenAI-compatible API server, 6 new messaging adapters, 4 new inference providers, MCP server management with OAuth 2.1, @ context references, gateway prompt caching, streaming enabled by default, and a sweeping reliability pass with 200+ bug fixes.
GitHub
V0.3.0
v0.3.0Mar 17, 20264 PRs / 0 issues
The streaming, plugins, and provider release — unified real-time token delivery, first-class plugin architecture, rebuilt provider system with Vercel AI Gateway, native Anthropic provider, smart approvals, live Chrome CDP browser connect, ACP IDE integration, Honcho memory, voice mode, persistent shell, and 50+ bug fixes across every platform.
GitHub
V0.2.0
v0.2.0Mar 12, 202643 PRs / 119 issues
First tagged release since v0.1.0 (the initial pre-public foundation). In just over two weeks, Hermes Agent went from a small internal project to a full-featured AI agent platform — thanks to an explosion of community contributions. This release covers **216 merged pull requests** from **63 contributors**, resolving **119 issues**.
GitHub

Why Hermes Agent?

Compared to OpenClaw and similar CLI-only agents

FeatureHermes Agent (73k stars)OpenClaw (356k stars)
Stars73k+356k+
LanguagePython — native ML/data science ecosystemTypeScript — Node.js 24 runtime
Channels12+ (Telegram, Discord, Slack, WhatsApp, Signal, Email, DingTalk, Matrix, Mattermost, Feishu, WeCom, SMS)24+ (all of Hermes channels plus iMessage, Google Chat, LINE, IRC, Teams, Nostr, Twitch, Zalo, WeChat, WebChat, etc.)
Inference providers400+ models via Nous Portal, Anthropic, OpenAI, HuggingFace, GitHub Copilot, Alibaba, and credential pool rotationMulti-provider with OAuth + API key auth, model failover chains
MCP supportNative MCP client + server mode (expose conversations to Claude Desktop, Cursor, etc.) with OAuth 2.1MCP client via tools
MemoryPluggable memory provider ABC — Honcho, vector stores, custom DBs with profile-scoped isolationSession model with main/group isolation, queue modes
Plugin systemPython plugin architecture with lifecycle hooks (pre/post LLM call), 70+ bundled skillsSkills system with bundled/managed/workspace skills, onboarding-driven setup
VoicePush-to-talk CLI, voice notes in Telegram/Discord, local Whisper transcriptionVoice Wake with wake words (macOS/iOS), Talk Mode on Android, ElevenLabs TTS
Companion appsCLI + gateway platformsmacOS menu bar app, iOS/Android nodes, Live Canvas with A2UI
Multi-instanceProfiles — isolated config, memory, sessions, skills per instance with token-lockMulti-agent routing — route channels/accounts to isolated agents per workspace
API serverOpenAI-compatible /v1/chat/completions endpoint + /api/jobs REST APIGateway WS control plane with sessions, presence, webhooks
BrowserCamoufox anti-detection browser with persistent sessions, VNC debugging, configurable SSRF bypassStandard browser tooling

Your AI Agent,
Everywhere

One agent for every platform.
CLI, Telegram, Discord, Slack, WhatsApp, and beyond.

github.com/NousResearch/hermes-agent