πŸš€ Hermes Windows Native Guide

May 8, 2026 Β· View on GitHub

License: MIT [Platform: Windows] [No Docker] [Docs: English]

πŸ‡ΊπŸ‡Έ English | πŸ‡¨πŸ‡³ δΈ­ζ–‡

AI Agent running natively on your Windows β€” No Docker Β· No WSL2 Β· Zero overhead


⚑ Quick Start β€” 3 Commands, 3 Minutes

# Step 1: Clone the repository
git clone https://github.com/markwang2658/hermes-windows-native.git
cd hermes-windows-native

# Step 2: One-click install
.\install.ps1

# Step 3: Start the service
.\start.ps1

Open http://127.0.0.1:8787 in your browser to get started.

πŸ“– Detailed Installation Instructions (click to expand)

Prerequisites

RequirementMinimum VersionCheck Command
Windows10 (1809+)[Environment]::OSVersion.VersionString
Python3.10+python --version
GitAny versiongit --version

Three Installation Methods

MethodBest ForTime Required
A. One-Click Script ⭐ RecommendedBeginners, quick start3 minutes
B. Manual InstallAdvanced users who want details10 minutes
C. Developer SetupContributors, code modification15 minutes

See: Agent Installation Guide | WebUI Installation Guide


πŸ› οΈ Tech Stack

TechnologyVersionPurpose
Python3.10+Runtime environment
PowerShell5.1+ / 7+Installation scripts / automation
Kimi K2.6LatestCloud chat model (Moonshot API)
GLM-4.6V-FlashQ4 quantizedLocal vision model (LM Studio)
faster-whisperlatestLocal speech-to-text
LM StudiolatestLocal model inference engine
CUDA12.x / 13.xGPU acceleration (optional)

πŸ“ Document Structure

docs/en/
β”œβ”€β”€ index.md                          # Main navigation (this file)
β”œβ”€β”€ installation/
β”‚   β”œβ”€β”€ agent-install.md              # Agent installation guide
β”‚   └── webui-install.md              # WebUI installation guide
β”œβ”€β”€ configuration/
β”‚   β”œβ”€β”€ chat-config.md                # Chat model config (Kimi K2.6)
β”‚   β”œβ”€β”€ vision-config.md              # Vision model config (GLM-4.6V)
β”‚   β”œβ”€β”€ audio-config.md               # Speech-to-text config (Whisper)
β”‚   └── text-config.md                # Text processing config (GLM-4.6V)
β”œβ”€β”€ architecture/
β”‚   └── trimode-routing.md            # Four-mode routing architecture
β”œβ”€β”€ troubleshooting/
β”‚   β”œβ”€β”€ index.md                      # Troubleshooting navigation
β”‚   β”œβ”€β”€ installation-issues.md        # Installation issues (7 items)
β”‚   β”œβ”€β”€ startup-issues.md             # Startup issues (6 items)
β”‚   β”œβ”€β”€ connection-issues.md          # Connection issues (4 items)
β”‚   β”œβ”€β”€ model-issues.md               # Model & API issues (6 items)
β”‚   β”œβ”€β”€ feature-issues.md             # Feature issues (6 items)
β”‚   β”œβ”€β”€ performance-issues.md         # Performance issues (5 items)
β”‚   └── windows-issues.md             # Windows-specific issues (8 items)
└── contributing.md                   # Contributing guide

Total: 16 documents, 42+ troubleshooting entries


πŸ“š Document Navigation

πŸ”§ Installation Guides (Required)

DocumentDescriptionEst. TimeDifficulty
Agent InstallationInstall Hermes Agent from scratch10 min⭐⭐
WebUI InstallationSet up browser interface5 min⭐

βš™οΈ Configuration Guides (As Needed)

DocumentDefault ModelUse CaseDifficulty
Chat Config (Kimi K2.6)Kimi K2.6 ☁️AI chat, code writing⭐
Vision Config (GLM-4.6V)GLM-4.6V πŸ”’Image recognition, screenshot analysis⭐⭐
Audio Config (Whisper)faster-whisper πŸ”’Voice message transcription⭐⭐
Text Config (GLM-4.6V)GLM-4.6V πŸ”’File summarization, code reading⭐⭐

☁️ = Cloud API (requires network) | πŸ”’ = Local inference (privacy-safe)

πŸ—οΈ Architecture Design (Advanced)

DocumentDescriptionDifficulty
Four-Mode Routing ArchitectureUnderstand how input routes to different models⭐⭐⭐

πŸ› Troubleshooting

DocumentDescription
Common Issues & Solutions42+ common problems and fixes

🀝 Contributing

DocumentDescription
Contributing GuideCommit convention / PR workflow / Checklist

New Users:       Agent Install β†’ WebUI Install β†’ Chat Config β†’ Start Using
Privacy-First:    Agent Install β†’ WebUI Install β†’ Vision/Audio/Text Config (all local)
Advanced Users:  Full documentation β†’ Architecture Design β†’ Custom Tuning
Contributors:     Contributing Guide β†’ Pick an Issue β†’ Fork β†’ Modify β†’ Submit PR

LinkDescription
πŸ“¦ Code Repositoryhermes-windows-native
🧠 Upstream AgentNousResearch/hermes-agent
🌐 Upstream WebUInesquena/hermes-webui
πŸ’¬ DiscussionsQ&A and discussions
πŸ› IssuesBug reports
🀝 Contributing GuideHow to contribute

πŸ“Š Project Highlights

FeatureDescription
πŸͺŸ Windows NativeNo Docker, no WSL2, no virtualization overhead
πŸ”’ Privacy FirstImages/audio/text processed locally β€” data never leaves your machine
πŸ’° Cost EffectiveChat uses cloud (affordable), everything else is free (local)
πŸ“š Comprehensive Docs16 documents + 42+ troubleshooting entries
🌐 Bilingual SupportFull internationalization (Chinese + English)

πŸ“„ License

This project is licensed under the MIT License.

Why MIT? Maximum community participation β€” allows both commercial and personal use.