Setup: DeepSeek + Hermes Agent

[Vlazoo]

DeepSeek + Hermes Agent Setup Plan

Overview

Install Hermes Agent (NousResearch's self-improving AI agent) configured to use DeepSeek as the model backend.

Prerequisites

• Linux/macOS/WSL2
• Python 3.8+
• DeepSeek API key (from https://platform.deepseek.com)
• DeepSeek account with credits (it's a paid API)

Installation Steps

Step 1: Install uv (fast Python package manager)

curl -LsSf https://astral.sh/uv/install.sh | sh

Step 2: Add ~/.local/bin to PATH

Add to ~/.bashrc (or ~/.zshrc for zsh):

echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc
source ~/.bashrc

Step 3: Install Hermes Agent

uv tool install hermes-agent

Step 4: Configure DeepSeek as provider

Create/edit ~/.hermes/config.yaml:

model:
  provider: deepseek
  model: deepseek-chat

Step 5: Set API key

Create ~/.hermes/.env:

echo 'DEEPSEEK_API_KEY=your-api-key-here' > ~/.hermes/.env

Note: Built-in providers (like deepseek) require the API key as an environment variable in .env, not in config.yaml.

Step 6: Verify installation

hermes --version
hermes chat -q "Hello, confirm you're DeepSeek"

Troubleshooting

• Run hermes doctor to diagnose config issues
• Error 401: Invalid API key
• Error 402: Insufficient balance — add credits at platform.deepseek.com

References

• Hermes Agent: https://github.com/nousresearch/hermes-agent
• Hermes Docs: https://hermes-agent.nousresearch.com/docs
• DeepSeek API: https://platform.deepseek.com
• Provider Config: https://hermes-agent.nousresearch.com/docs/integrations/providers