Self-hosted AI agent with persistent memory, scheduled jobs, and messaging-app access. Replaces ChatGPT Plus + Zapier.
Hermes remembers who you are, your stack, your conventions, and every correction you've made. No CLAUDE.md to maintain — memory accumulates automatically as a side effect of normal use.
Cron tasks fire on your server while you're offline. Morning briefings, PR reviews, test runs, datasource monitoring — Hermes wakes up, runs the task with full memory access, and delivers the result wherever you want.
Same agent, same memory, same history — accessible from a terminal, the web UI, Telegram, Discord, Slack, WhatsApp, Signal, SMS, or email. Start on your phone, finish on your laptop.
When Hermes solves a tricky problem, it saves the approach as a reusable skill. No marketplace to browse, no plugins to install — skills emerge from your actual workflow.
Bring your own keys for OpenAI, Anthropic, Google, DeepSeek, OpenRouter — or pay-as-you-go through Vendo's metered proxies. Swap models any time without losing memory.
Need heavy lifting? Hermes can spawn Claude Code or Codex for complex tasks and bring the results back into its own memory. Long-horizon workflows without losing context.


Hermes runs on your own deployment. You pay Vendo for the hosting and for any LLM tokens or messaging your agent burns through — billed per-use against your credit balance, no monthly minimums.
Pay as you go — no subscriptions, no per-seat fees. Every model we bill through is listed in our rate card.