Set Up Hermes Agent with DeepSeek

Run a self-improving AI agent 24/7 for pocket change. Install Hermes Agent, point it at DeepSeek, and talk to it from your phone over Telegram.

๐Ÿ’ธ

Why Hermes Agent + DeepSeek?

Hermes Agent by Nous Research is an open-source agent that keeps memory, writes its own reusable skills, runs scheduled jobs and answers you on chat apps. Because it runs all day and re-sends a long system prompt, memory and skill context on every turn, most of what it spends goes on input tokens, and most of those repeat from turn to turn.

DeepSeek caches repeated prompt prefixes automatically and bills cache hits at a tiny fraction of the normal input price. For a long-running agent that combination of a large stable context and cheap cache hits matters most. A busy personal agent can cost less per month than one lunch.

$0.003
per 1M cached input tokens on deepseek-flash (off-peak; $0.006 peak)
1M
token context window on both DeepSeek Flash and V4 Pro
~100ร—
cheaper per cached token than typical frontier APIs ($0.30โ€“$0.50 / 1M)
๐Ÿงฎ

Monthly Agent Cost Estimator

Drag the sliders to match your agent's workload.

Agents re-send the same system prompt, memory and skills each turn, so 80โ€“95% is common.
Agent replies and tool calls are short compared with the context they read.
Peak = 01:00โ€“04:00 and 06:00โ€“10:00 UTC on weekdays (2ร— price). "Blended" assumes usage spread evenly across the week (~21% peak).

DeepSeek rates from the official pricing page (November 2026). Frontier rows use representative list prices for comparison: mid-tier frontier at $3 input / $0.30 cached / $15 output per 1M, and flagship frontier at $5 / $0.50 / $25. Real bills vary with model choice, prompt structure and tool usage, so treat these numbers as estimates.

๐Ÿ—บ๏ธ

The Setup Roadmap

Three steps, about 30 minutes in total. Each step has a quick summary. Click its button to reveal the full, detailed guide.

1 Set up Hermes AgentInstall on Docker, macOS or Windows and keep it running 24/7.
๐Ÿณ Docker (recommended)๐ŸŽ macOS ยท Homebrew + uv + launchd๐ŸชŸ Windows ยท WinGet + PowerShell
  1. Docker: create ~/.hermes, run the one-time setup wizard in a throwaway container, then start gateway run with --restart unless-stopped.
  2. macOS: brew install uv git ripgrep ffmpeg node, run the official install script, then keep the gateway alive with a launchd agent.
  3. Windows: install Git and uv with WinGet, run the PowerShell installer, and start the gateway from a terminal or Task Scheduler.
  4. Check everything with hermes doctor.
mkdir -p ~/.hermes
docker run -it --rm -v ~/.hermes:/opt/data nousresearch/hermes-agent setup
2 Connect DeepSeek providersUse the direct DeepSeek API, the OpenCode gateway, or Command Code.
๐Ÿ”‘ Direct DeepSeek API๐Ÿงฉ OpenCode Zen / GoโŒจ๏ธ Command Code
  1. Direct API (cheapest per token): add DEEPSEEK_API_KEY to ~/.hermes/.env and choose DeepSeek in hermes model.
  2. OpenCode (flat subscription): set OPENCODE_GO_API_KEY or OPENCODE_ZEN_API_KEY and pick the opencode-go or opencode-zen provider.
  3. Command Code: set COMMANDCODE_API_KEY (provider commandcode) to reuse your plan in Hermes, and use the cmd CLI for coding work.
  4. Pick deepseek-flash for everyday turns and deepseek-v4-pro for harder reasoning.
echo 'DEEPSEEK_API_KEY=sk-your-key' >> ~/.hermes/.env
hermes model   # choose DeepSeek โ†’ deepseek-flash
3 Connect Telegram & mobile clientsChat with your agent from iOS or Android, locked down to your user ID.
๐Ÿค– @BotFather๐Ÿ†” @userinfobot whitelist๐Ÿ“ฑ iOS & Android
  1. Message @BotFather, send /newbot, and copy the bot token.
  2. Message @userinfobot to get your numeric Telegram user ID.
  3. Put TELEGRAM_BOT_TOKEN and TELEGRAM_ALLOWED_USERS in ~/.hermes/.env, or run hermes gateway setup.
  4. Restart the gateway and send your bot a message from your phone. Voice notes are transcribed automatically.
TELEGRAM_BOT_TOKEN=123456789:AA...your-token
TELEGRAM_ALLOWED_USERS=987654321
Result: an always-on agent on your own hardware or a cheap VPS, running on DeepSeek, that you can message from anywhere and that remembers context between conversations.