Self-Hosting Hermes Agent on Netcup VPS: Complete Guide to the Self-Improving AI Agent
TL;DR: Running Hermes Agent in 5 Minutes
1. Get a Netcup VPS 1000 G12 (4 vCore, 8 GB RAM, 256 GB NVMe SSD — €10.37/month)
Use coupon 5799nc17755735170 for 1 month free.
2. SSH into your server and run the installer:
ssh root@your-server-ip
curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash
source ~/.bashrc
hermes setup
hermes
3. Done! Configure your API keys and start chatting with your autonomous AI agent.
Need more power? Check the full server comparison below — including Root Servers with dedicated CPU cores and up to 2 months free with coupon codes.
Introduction: The Rise of Self-Improving Autonomous AI Agents
The AI landscape has dramatically evolved from simple chatbots that respond to prompts within fixed boundaries to autonomous agents capable of initiating actions, making decisions, and adapting without constant human oversight. Hermes Agent represents this paradigm shift — it's not just another AI chatbot but a self-improving autonomous agent that grows more capable the longer it runs.
Created by Nous Research, Hermes Agent has accumulated over 108,000 GitHub stars in just seven weeks since its February 2026 release, making it the fastest-growing AI agent framework of 2026. Unlike traditional AI assistants, Hermes Agent features a built-in learning loop that automatically creates reusable skills from experience, refines those skills during continued use, and builds a deepening model of who you are across sessions.
To unlock the full potential of autonomous AI agents like Hermes Agent, you need reliable infrastructure that's always available — and that's where VPS hosting comes in. Running Hermes Agent on a Netcup VPS gives you a powerful, always-on AI assistant at a fraction of the cost of major cloud providers. This guide covers everything from what Hermes Agent is, to choosing the right Netcup server, deploying it, and running it securely in production.
What is Hermes Agent?
Hermes Agent is an open-source autonomous AI agent built by Nous Research — the AI research lab behind Hermes, Nomos, and Psyche. Its tagline — "The agent that grows with you" — perfectly captures its core design philosophy: unlike chatbots that reset after each conversation, Hermes Agent learns, remembers, and improves over time.
Key Features of Hermes Agent
Closed Learning Loop
Hermes Agent is the only AI agent with a built-in learning loop. It automatically generates reusable skills from experience, refines those skills during continued use, nudges itself to persist knowledge, searches its own past conversations, and builds a deepening model of your preferences across sessions. This means the longer you use it, the more capable it becomes.
Multi-Platform Support
Hermes Agent lives where you do — Telegram, Discord, Slack, WhatsApp, Signal, Email, CLI, Matrix, Mattermost, DingTalk, Feishu, WeCom, and Home Assistant — all from a single gateway process. Start a conversation on one platform and continue on another without losing context.
Flexible Model Selection
Use any LLM you want — Nous Portal, OpenRouter (200+ models), NVIDIA NIM (Nemotron), Xiaomi MiMo, z.ai/GLM, Kimi/Moonshot, MiniMax, Hugging Face, OpenAI, Anthropic, or your own endpoint. Switch models with a single command without code changes.
Persistent Memory and Skills
Hermes Agent maintains bounded, curated memory that persists across sessions. It remembers your preferences, project contexts, and past solutions. The skills system allows on-demand knowledge documents that the agent can load when needed, and it can autonomously create new skills after complex tasks.
Scheduled Automations
Built-in cron scheduler with natural language expressions. Set up daily reports, nightly backups, weekly audits — all running unattended through the messaging gateway.
Full Web and Browser Control
Web search, browser automation, vision capabilities, image generation, text-to-speech, and multi-model reasoning are all included out of the box.
118 Pre-Built Skills
Version 0.10.0 ships with 118 pre-built skills covering developer workflows, research tasks, writing, data processing, and system administration.
Why Self-Hosting Hermes Agent Matters
Data Privacy and Control
Running Hermes Agent on your own VPS means your conversations, data, and AI interactions stay under your control. This is critical for enterprises handling sensitive information or individuals concerned about privacy.
Cost Efficiency
Major cloud AI services charge per-request fees that add up quickly. Self-hosting Hermes Agent on a VPS gives you unlimited interactions for a flat monthly fee — often 60-80% cheaper than cloud alternatives.
Always-On Availability
Your laptop sleeps. Your home internet drops. A VPS in a data center with redundant power and networking keeps Hermes Agent available 24/7 from any device, anywhere in the world.
Customization and Extensibility
Self-hosting gives you full access to configure, customize, and extend Hermes Agent according to your specific needs — something cloud services don't always allow.
How to Use Hermes Agent
Installation
The installation process is remarkably straightforward:
curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash
source ~/.bashrc
The installer automatically handles all dependencies — Python, Node.js, ripgrep, ffmpeg, and the virtual environment setup.
Initial Setup
After installation, run the setup wizard to configure your AI assistant:
hermes setup
This will guide you through:
- Choosing your LLM provider and model
- Configuring enabled tools
- Setting up the messaging gateway (optional)
- Configuring API keys
Basic Usage
Start an interactive conversation:
hermes
Change the model:
hermes model
# or
/model provider:model
Configure tools:
hermes tools
Set up messaging platforms:
hermes gateway setup
hermes gateway start
Check system health:
hermes doctor
Messaging Platforms
Once the gateway is configured, you can interact with Hermes Agent through:
- Telegram: Start a chat with your bot username
- Discord: Use the bot in your server
- Slack: Message the bot directly
- WhatsApp: Link your phone number
- Signal: Connect your Signal account
- Email: Send emails to your bot
Key Commands
| Action | CLI Command | Messaging Command |
|---|---|---|
| Start chatting | hermes |
Send a message |
| New conversation | /new or /reset |
/new or /reset |
| Change model | hermes model |
/model [provider:model] |
| Check usage | /usage |
/usage |
| Browse skills | /skills |
/skills |
| Interrupt work | Ctrl+C | /stop |
Quick Start Guide: Deploy Hermes Agent on Netcup VPS
Step 1: Create a Netcup Account
Visit netcup.com/en and sign up. Verification takes just a few minutes. Use one of our coupon codes from this guide to get your first month free.
Step 2: Order Your Server
Choose a VPS or Root Server plan based on the recommendations in the next section. For getting started, we recommend VPS 1000 G12 — it provides ample resources for personal use and development.
Step 3: Select Operating System
During setup, select Ubuntu Server 22.04 LTS or Debian 12 as the operating system. Both are fully supported.
Step 4: Configure Access
- SSH Keys (recommended): Upload your public key through the control panel
- Hostname: Set something meaningful (e.g.,
hermes.yourdomain.com) - Root Password: Use at least 12 characters with mixed case, numbers, and special characters
Step 5: Choose Data Center Location
Select the location closest to you for lowest latency. Available locations include Vienna (VIE), Nuremberg (NUE), Amsterdam (AMS), and Manassas/USA (MNZ).
Step 6: Wait for Deployment
Netcup typically provisions servers in 5-15 minutes. You'll receive an email when your server is ready.
Step 7: Connect via SSH
ssh root@your-server-ip
# Or with SSH key
ssh -i /path/to/your/key.pem root@your-server-ip
Step 8: Install Hermes Agent
curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash
source ~/.bashrc
Step 9: Run Setup Wizard
hermes setup
Follow the prompts to:
- Choose your LLM provider (OpenRouter recommended for flexibility)
- Enter your API key
- Configure tools (start with defaults)
- Optionally set up the messaging gateway
Step 10: Start Using Hermes Agent
hermes
Type your first message and watch your self-improving AI assistant respond!
Step 11: Configure Firewall (Recommended)
ufw allow 22/tcp
ufw allow 80/tcp
ufw allow 443/tcp
ufw enable
Step 12: Set Up Gateway (Optional)
For messaging platform access:
hermes gateway setup
hermes gateway start
systemctl enable hermes-gateway
Choosing the Right Netcup Server for Hermes Agent
Netcup offers several server product lines. Here's how each fits Hermes Agent use cases, with coupon codes for each.
VPS (Virtual Private Server) — Best Value for Most Users
Netcup's VPS line uses shared hardware with guaranteed resource minimums. Ideal for personal use, development, and small-to-medium production workloads.
VPS 500 G12 — Light Usage
| Resource | Specification |
|---|---|
| vCPU Cores | 2 |
| RAM | 4 GB DDR5 ECC |
| Storage | 128 GB NVMe SSD |
| Bandwidth | Unlimited |
| Monthly Cost | €5.91 |
Best For:
- Testing Hermes Agent
- Light personal use
- Development environments
- Single-user setups with minimal concurrent tasks
Coupon: Use coupon 5799nc17755734540 for 1 month free.
VPS 1000 G12 — Recommended Starting Point
| Resource | Specification |
|---|---|
| vCPU Cores | 4 |
| RAM | 8 GB DDR5 ECC |
| Storage | 256 GB NVMe SSD |
| Bandwidth | Unlimited |
| Monthly Cost | €10.37 |
Best For:
- Personal Hermes Agent instances
- Production single-user setups
- Development and testing with real workloads
- Running gateway for 1-2 messaging platforms
Coupon: Use coupon 5799nc17755732650 for 1 month free.
VPS 2000 G12 — Recommended for Production
| Resource | Specification |
|---|---|
| vCPU Cores | 8 |
| RAM | 16 GB DDR5 ECC |
| Storage | 512 GB NVMe SSD |
| Bandwidth | Unlimited |
| Monthly Cost | €19.25 |
Best For:
- Production Hermes Agent deployments
- Running multiple concurrent AI tasks
- Heavy tool usage (browser automation, web scraping)
- Teams of 2-5 users
- Multiple messaging platforms
Coupon: Use coupon 5800nc17718015234 for 1 month free.
VPS 4000 G12 — Heavy Workloads
| Resource | Specification |
|---|---|
| vCPU Cores | 12 |
| RAM | 32 GB DDR5 ECC |
| Storage | 1 TB NVMe SSD |
| Bandwidth | Unlimited |
| Monthly Cost | €32.41 |
Best For:
- High-throughput AI processing
- Large teams or multi-tenant setups
- Running Hermes Agent alongside other services
- Complex autonomous workflows with heavy data processing
Coupon: Use coupon 5801nc17755880762 for 1 month free.
VPS 8000 G12 — Maximum Power
| Resource | Specification |
|---|---|
| vCPU Cores | 16 |
| RAM | 64 GB DDR5 ECC |
| Storage | 2 TB NVMe SSD |
| Bandwidth | Unlimited |
| Monthly Cost | €47.95 |
Best For:
- Enterprise-grade deployments
- Maximum concurrent users
- Running multiple AI agents
- Compute-intensive AI tasks
Coupon: Use coupon 5802nc17718015175 for 1 month free.
Root Servers (RS) — Dedicated Performance
Root Servers provide dedicated CPU cores (not shared), giving you consistent performance without noisy-neighbor effects. Choose these when predictable performance matters more than cost.
RS 1000 G12 — Entry Dedicated
| Resource | Specification |
|---|---|
| CPU Cores | 4 (dedicated) |
| RAM | 8 GB DDR5 ECC |
| Storage | 256 GB NVMe SSD |
| Bandwidth | Unlimited |
| Monthly Cost | €12.78 |
Best For:
- Production Hermes Agent with guaranteed performance
- Latency-sensitive AI processing
- Workloads needing consistent CPU availability
Coupon: Use coupon 5997nc17755880772 for 2 months free.
RS 2000 G12 — Production Dedicated
| Resource | Specification |
|---|---|
| CPU Cores | 8 (dedicated) |
| RAM | 16 GB DDR5 ECC |
| Storage | 512 GB NVMe SSD |
| Bandwidth | Unlimited |
| Monthly Cost | €22.96 |
Best For:
- Teams running Hermes Agent in production
- AI workloads requiring sustained CPU performance
- Running multiple services alongside Hermes
Coupon: Use coupon 5160nc17718015412 for 2 months free.
RS 4000 G12 — Enterprise Dedicated
| Resource | Specification |
|---|---|
| CPU Cores | 12 (dedicated) |
| RAM | 32 GB DDR5 ECC |
| Storage | 1 TB NVMe SSD |
| Bandwidth | Unlimited |
| Monthly Cost | €41.62 |
Best For:
- Enterprise-grade Hermes Agent deployments
- Heavy multi-user AI workloads
- Maximum performance without resource contention
Coupon: Use coupon 5161nc17718015393 for 2 months free.
RS 8000 G12 — Maximum Dedicated Power
| Resource | Specification |
|---|---|
| CPU Cores | 16 (dedicated) |
| RAM | 64 GB DDR5 ECC |
| Storage | 2 TB NVMe SSD |
| Bandwidth | Unlimited |
| Monthly Cost | €71.82 |
Best For:
- Large-scale AI operations
- Running multiple Hermes Agent instances
- Organizations needing maximum compute
Coupon: Use coupon 5162nc17718015365 for 2 months free.
VPS vs Root Server: Which Should You Choose?
| Factor | VPS | Root Server |
|---|---|---|
| CPU | Shared (burst capable) | Dedicated (consistent) |
| Price | Lower | Slightly higher |
| Performance | Great for variable workloads | Best for sustained workloads |
| Guarantee | 99.9% annual availability | 99.9% annual availability |
| Best for | Development, personal use, light production | Production, teams, latency-sensitive AI |
Our recommendation: Start with a VPS 1000 G12 using coupon 5799nc17718015261 to test your setup. Once you're ready for production, upgrade to VPS 2000 G12 or move to an RS 1000 G12 for dedicated performance.
Cost Comparison: Netcup vs Major Cloud Providers
Monthly Cost (approximate, excluding VAT)
| Plan | Netcup | AWS | Google Cloud |
|---|---|---|---|
| 4 vCPU, 8 GB RAM | €10.37 (VPS) | €25-30 | €30-35 |
| 8 vCPU, 16 GB RAM | €19.25 (VPS) | €55-65 | €60-70 |
| 12 vCPU, 32 GB RAM | €32.41 (VPS) | €110-130 | €120-140 |
| 16 CPU, 64 GB RAM | €47.95 (VPS) | €200+ | €220+ |
Annual Savings Example (Mid-Tier)
- Netcup VPS 2000: €231/year
- AWS equivalent: €720/year
- Annual savings: €489
Total Cost of Ownership
Beyond monthly fees, Netcup saves you:
- Setup time: Straightforward installation saves hours compared to cloud service configuration
- Management overhead: Intuitive control panel reduces admin time
- Support costs: Included support (no premium tiers required)
- Hidden fees: No surprise charges for bandwidth, backups, or basic support
- Scaling costs: Linear, predictable pricing with easy upgrades
Why Netcup for AI Agents?
Price-to-Performance Leader Netcup consistently delivers 60-80% cost savings compared to AWS, Google Cloud, or Azure for equivalent resources. For always-on AI agents, this adds up fast.
German Data Privacy (GDPR) Netcup operates under strict German and EU data protection laws. All data centers are in Germany and Austria, giving you strong data sovereignty and privacy guarantees — critical for AI applications handling sensitive data.
Enterprise Infrastructure Redundant power, multi-provider network connections, and advanced cooling in Netcup's data centers ensure the reliability AI workloads demand.
Transparent Pricing No introductory bait-and-switch. What you pay month one is what you pay ongoing. Bandwidth, backups, and support are included.
Easy Scaling Upgrade CPU, RAM, and storage through the control panel with near-instant effect. Start small, scale as your needs grow.
Security Best Practices
SSH Hardening
Disable password authentication and use SSH keys only:
# /etc/ssh/sshd_config
PasswordAuthentication no
PubkeyAuthentication yes
Avoid Root for Daily Use
adduser hermes-admin
usermod -aG sudo hermes-admin
API Key Security
Never expose Hermes Agent API endpoints without authentication. Use API keys or OAuth tokens for all external access.
SSL/TLS
Encrypt all traffic. Use Let's Encrypt for free certificates:
apt install certbot python3-certbot-nginx
certbot --nginx -d hermes.yourdomain.com
Firewall
Only open ports you actually need. Use both Netcup's built-in firewall and system-level ufw:
ufw allow 22/tcp
ufw allow 443/tcp
ufw enable
Resource Limits
Prevent runaway processes from consuming all resources:
# /etc/systemd/system/hermes.service
[Service]
MemoryLimit=4G
CPUQuota=200%
Keep Everything Updated
apt update && apt upgrade
hermes update
Logging and Alerting
Monitor logs for issues:
journalctl -u hermes -f
Set up alerts for failed authentication attempts, resource spikes, and error rate increases.
GDPR Compliance
Netcup's European data centers provide a strong foundation for compliance. On your end: minimize data collection, implement retention policies, and use anonymization where possible.
Real-World Use Cases
Personal AI Assistant
A traveling professional accesses Hermes Agent from laptops, tablets, and phones. All AI processing happens on the Netcup VPS, so even lightweight devices get full AI capabilities. Complete interaction history stays synced regardless of device.
Development Workflow Automation
A software development team uses Hermes Agent to automatically review pull requests, identify bugs, suggest improvements, and generate documentation. Hosted on Netcup, it's available 24/7 across time zones and keeps sensitive code isolated from developer machines.
Automated Customer Service
An e-commerce company deploys Hermes Agent as an intelligent customer service agent. Netcup's European infrastructure ensures GDPR compliance for customer data. Resources scale during holiday peaks, and the flat monthly cost beats per-request pricing from enterprise AI platforms.
Research Environment
A university research group shares a single Hermes Agent instance for dataset analysis, literature reviews, and AI experiments. Netcup's pricing maximizes computing resources per grant dollar, and the shared environment ensures reproducible results.
DevOps Automation
A DevOps team uses Hermes Agent to monitor infrastructure, deploy applications, and respond to alerts across multiple cloud providers. Hosting on a reliable Netcup VPS ensures the automation engine itself stays online with enterprise-grade uptime.
Content Generation
A media company uses Hermes Agent for content summaries, article curation, and draft generation across languages. Self-hosted on Netcup, they get unlimited generation for a flat fee with full control over style and vocabulary tuning.
Migration Guide: Local Installation to Netcup VPS
If you're running Hermes Agent locally and want to move to Netcup, follow these steps.
1. Document Your Current Setup
Note your Hermes Agent version, configuration files, dependencies, data directories, custom skills, and API keys.
2. Backup Everything
# Configuration
tar -czf hermes-config-backup.tar.gz ~/.hermes
# Skills
tar -czf hermes-skills-backup.tar.gz ~/.hermes/skills
# Memory
tar -czf hermes-memory-backup.tar.gz ~/.hermes/memory
3. Deploy Your Netcup Server
Order a VPS or Root Server, select your preferred OS, and wait for provisioning.
4. Install Hermes Agent
ssh root@your-netcup-ip
curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash
source ~/.bashrc
hermes setup
5. Transfer Data
# From your local machine
scp hermes-config-backup.tar.gz root@your-netcup-ip:/tmp/
scp hermes-skills-backup.tar.gz root@your-netcup-ip:/tmp/
scp hermes-memory-backup.tar.gz root@your-netcup-ip:/tmp/
# On the server
cd /tmp
tar -xzf hermes-config-backup.tar.gz -C ~/
tar -xzf hermes-skills-backup.tar.gz -C ~/
tar -xzf hermes-memory-backup.tar.gz -C ~/
6. Update Configuration
- Replace
localhostreferences with your VPS hostname/IP - Update API keys if they're tied to specific IPs
- Configure firewall rules for Hermes Agent's ports
7. Test
hermes doctor
8. Go Live
Update DNS records, set up SSL with Let's Encrypt, and update client configurations.
9. Post-Migration
- Monitor performance for the first week
- Keep your local installation as a fallback for at least 7 days
- Fix any permission issues:
chown -R $USER:$USER ~/.hermes - Check logs if anything fails:
journalctl -u hermes -xe
Conclusion
Hermes Agent on Netcup is a powerful, cost-effective combination for running autonomous self-improving AI agents. Netcup's infrastructure delivers the reliability, performance, and privacy that AI workloads require — at prices 60-80% below major cloud providers.
To get started:
- Pick a server plan from the recommendations above
- Use a coupon code from this guide for free month(s)
- Deploy with the quick install script
- Configure your LLM provider and API keys
- Start building with your always-on, self-improving AI assistant
Whether you're a solo developer, a growing team, or an enterprise, there's a Netcup plan that fits. Start with a VPS 1000 G12 using coupon 5799nc17755868250 and scale from there.
This guide is intended to help you understand and deploy Hermes Agent on Netcup. Technology changes rapidly — always refer to the official Hermes Agent and Netcup documentation for the most current information.