Hermes Agent by Nous Research is an open-source agent that learns skills as it goes, run on your own machine across five sandboxes, or hosted on Hermes Cloud. Rerun gives your team agents that learn too, hosted together on one private machine, with the apps and the board around them.
Last reviewed , against each product’s public pricing and docs.
Pick Hermes Agent if you want MIT-licensed code, a choice of five backends from local to Modal, and full control of the runtime, or one agent on Hermes Cloud that you reach from Telegram, Discord or WhatsApp. Pick Rerun if you want several agents for a team on one machine included in a flat plan, 3,000+ apps connected from the browser, and ready-made agents to start from.
Side by side
Hermes is one agent you install and run, or host on Hermes Cloud by the day. Rerun gives you a team of agents already doing the job, on one machine included in the plan.
Your computer or server, across five backends you pick from, or Hermes Cloud, where each agent gets its own dedicated instance.Free to self-host. Hermes Cloud wants $2 of credit before it deploys, then bills the instance by the day, metered to the second, with inference and tools on top. Nous Portal plans are $20, $100 or $200 a month of credit.Any MCP server, configured and filtered by you.Telegram, Discord, Slack, WhatsApp, Signal, email or the CLI.A blank agent that writes its own skills as it goes.Key differences
Four differences that change your day to day, shown on the real product.
Hermes runs where you install it, or on Hermes Cloud, where each agent gets its own dedicated instance billed by the day. On Rerun, one private machine per workspace runs all your agents, and it is included in the plan.
A payment just failed
Margo is chasing it
New email from a lead
Jules is drafting a reply
Every morning at 7:00
Sam sorted your inbox
Hermes stores conversations, memories and skills between sessions and writes skills on its own. Rerun agents do too, and you see what they learned and every action they took on a live board, next to one company brain the whole workspace shares.
## Procedure
1. Pull unpaid invoices from Stripe daily
2. Send a first reminder on day 7 past due
3. Escalate to the account owner after 2 tries
4. Log every touch in the CRM thread
## Exceptions
- Never chase an invoice under $50
- Read references/tone.md before escalating
## Billing
- Pays around day 30, sometimes later
- A PO number is required on every invoice
- Invoices go to ap@acme.co, not to Dana
## People
- Dana Reyes runs ops, answers within a day
- Escalations go through their COO
## Voice
- English, formal, no emoji
- Sign off as the founder, not an assistant
- Keep it under 120 words
## Timing
- Follow up twice, then stop
- Send whenever the draft is ready
Agents stop before anything consequential and ask in the app, seats are free, and any teammate can answer. With Autopilot on, the Foreman answers within your rules while you are away, never for spending or deleting.
Hermes reaches any MCP server you configure and filter yourself. On Rerun, 3,000+ apps connect from the browser with a guided sign-in, any MCP server still works, and each agent runs on the Rerun model included in the plan, your ChatGPT or Claude plan, or your own key.
Switching
No migration project. Start with the one task that costs you the most, and let it prove itself.
Give the agent the job in plain words. It writes its own instructions and skills.
Keep the model provider you like, and sign in to your tools from the browser.
The agent now runs on your private machine with its own schedule, and you stop paying for or maintaining the instance Hermes ran on.
Pricing
Hermes Agent is free and MIT-licensed to run yourself, and you pay for the model and the machine. Hermes Cloud wants at least $2 of credit before it deploys a Cloud Agent, then bills each instance by the day, metered to the second of uptime plus storage, with inference and tool usage billed on top; its instance sizes and prices are only listed inside the Nous Portal, whose plans are $20, $100 or $200 a month of credit. Rerun is one plan from $29 a month with a secured private machine, 5 agents, unlimited seats and model credit included.
Everything you need to put Coworkers to work
Add Coworkers or credits whenever you need them.
5-day free trial · card required, $0 today
5 Coworkers
Five Coworkers running side by side, each with its own job and tools.
Your own machine
A private computer in the cloud where every Coworker of your workspace lives, shares its files and its data. Only yours.
Unlimited seats
Your whole team on the same workspace, with no per-seat fee.
Bring your own AI
Optional: your own key, subscription or local model.
3,000+ apps
All your tools ready to connect, plus any MCP or API.
Delegate any task
Hand off the busywork. Your Coworkers run it around the clock.
Invite your teammates
Share your workspace and build your Coworkers together.
Further reading
The Best Hermes AI Alternative in 2026: Governed, Human-in-the-Loop AI AgentsLooking for a Hermes AI alternative? We rank and compare Hermes Agent, OpenClaw forks, Claude Code, Lindy and Rerun on security, control, and governance, and explain why a human-in-the-loop agent wins.
Hermes AI Pricing: The True Cost of Running Hermès Agent in 2026Hermès Agent is free and open-source, so why do real bills range from $8 to $200+ a month? A full breakdown of hosting, tokens, and the hidden ops costs, plus the transparent managed alternative.
Hermes vs OpenClaw (2026): Honest Comparison and When to Skip Self-HostingAn honest, head-to-head comparison of Hermes Agent and OpenClaw across architecture, memory, security, setup, and cost, plus when self-hosting either agent is the wrong call.Other comparisons
Paperclip gives the agents you already run an org chart, budgets and approval gates, on a server you host. Rerun gives you the agents themselves, already running on a machine we keep up.
Read the comparisonOpenClaw is an assistant you install, host and secure, and its foundation sells nothing. Rerun agents do the same jobs from an isolated machine we run for you.
Read the comparisonClaude takes on the tasks you hand it, on Claude models. Rerun runs your recurring work with a team of agents, on any model, your Claude plan included.
Read the comparison
Manus delivers one-off research, slides and sites, billed in credits. Rerun runs your recurring work every day, for one flat price.
Read the comparison
Grok Bot comes inside SuperGrok and Cursor plans, in beta. Rerun is a plan of its own, with agents ready to install on the model you choose.
Read the comparisonA Zap moves data between apps. A Rerun agent sorts the inbox, answers the lead and chases the invoice.
Read the comparisonA Make scenario passes data from module to module. A Rerun agent answers the lead, chases the invoice and learns from your corrections.
Read the comparisonn8n gives engineers workflows to build and host. Rerun gives your team agents that answer leads and chase invoices, with nothing to host.
Read the comparison
Lindy bills every active user, then every task in credits. Rerun runs the same jobs for one flat price, on the AI you already pay for.
Read the comparisonTwin is flat-priced too, but the browser agent and the lead data sit on its $149 plan, and the people you invite are clients, not teammates.
Read the comparison
Dust answers from your company’s documents, priced per seat. Rerun agents do the recurring work for the whole team, at one price.
Read the comparisonSquad is $99 a month on top of the AI plan you bring. Rerun costs less, includes model credit and has agents ready to install.
Read the comparisonFAQ
Yes. Nous hosts Hermes too, one dedicated instance per agent on Hermes Cloud. Rerun hosts a team of agents on one private machine per workspace, with 3,000+ apps and a board your whole team uses.
Yes. Rerun agents build skills and keep memory between runs, and they get sharper each time you correct them. You can read what they learned from the board.
The software is free. Self-hosted, you pay for the model and the machine. On Hermes Cloud, you need $2 of credit to deploy, each running instance is billed by the day and metered to the second, and inference and tools are billed on top; Nous Portal plans run $20, $100 or $200 a month. Rerun bundles the machine, 5 agents, unlimited seats and model credit into one plan from $29 a month.
Rerun runs on Claude, ChatGPT and more, and an OpenRouter key, including a custom endpoint, reaches other models.
No. You create and manage agents in the browser.
2,000+ businesses run on Rerun
Compare
Rerun vs Claude CoworkRerun vs ManusRerun vs Grok BotRerun vs PaperclipRerun vs Hermes AgentRerun vs OpenClawRerun vs ZapierRerun vs MakeRerun vs n8nRerun vs LindyRerun vs TwinRerun vs DustRerun vs SquadGuides
No-code AI agent buildersn8n alternativesPersonal AI agentsHermes vs OpenClawHermes alternativesClaude Code vs Codex© 2026 Rerun by Reunit SA. All rights reserved.
