NEW Telegram · Lark · Discord · WeChat — your assistant lives there in seconds

Your AI assistant. On the apps you already use.

Name: OpenClaw by RunClaw
Price: 19 USD

No server to manage. No API key to buy. Pick a chat app, pick a model, start chatting.

What you get, the moment you log in

A managed assistant with channels, model, and capabilities ready to wire up — no infrastructure on your plate.

Hello, your assistant · online

Pick the weight class that fits

Same channels, same models, same managed platform. OpenClaw runs heavier with more tools out of the box; Hermes runs lighter and learns from every task.

Full power

OpenClaw

A full-toolkit AI assistant from the OpenClaw open-source project. Browser, canvas, voice, multi-agent routing.

Browser + canvas + voice — first-class tools
Multi-agent routing for parallel work
Skill marketplace and native console chat
2 vCPU · 4 GB RAM · 40 GB SSD

From $15.8/mo

Learn more →

Lightweight

Hermes

The agent that grows with you, from Nous Research. Self-improving, learns from every task.

Built-in learning loop — writes its own skills from experience
Spawns isolated subagents for parallel workstreams
Built-in cron + scheduled automations
2 vCPU · 2 GB RAM · 20 GB SSD

From $4.2/mo

Learn more →

Connect Your Favorite Platforms

All major messaging platforms and AI models. Deploy once, reach everywhere.

Discord

WeChat

Feishu

WeCom

DingTalk

Models included. Free on every plan.

No API keys, no provider sign-ups. Just log in and run.

Kimi K2.6 · 256k ctx · vision

free

GLM-4.7 Flash · 128k ctx

free

One model pool, shared across every plan. No token metering. Quota resets daily.

Do the math: self-host vs RunClaw

From setup time to monthly cost — the gap, at a glance.

Self-host

Time to live ~35 min

Server / VPS

$5 – $10 / mo

Keep it running 24/7
AI subscription or API key

$20 – $100+ / mo

Starts at ChatGPT Plus; API metered by token
Ops time

On you

Deploy, update, debug, handle outages

Monthly cost $25 – $100+ / mo Your time not included. When it breaks, it's on you.

RunClaw

Time to live ~2 min

Runtime

Fully hosted

No laptop, no VPS required
Built-in models

Free on every plan

Kimi · GLM, no token metering
Ops

On us

Servers, security, updates — all handled

Starts at from $4.2 / mo Hermes from $4.2 · OpenClaw from $15.8 · 7-day free trial

Comparison based on public VPS and AI subscription pricing (2026-04). See each provider's site for current rates. RunClaw plans include runtime and built-in models.

Frequently Asked Questions

How long does it take to deploy an AI agent?

About 2 minutes. Sign up, pick a plan, connect your chat app, and your agent is live. No server setup or API keys needed.

Do I need to buy AI model API keys?

No. Both OpenClaw and Hermes include AI models out of the box — Kimi K2.6, GLM-4.7, and more. You can also bring your own models if you prefer.

What's the difference between OpenClaw and Hermes?

Both are fully managed and work out of the box with AI models included. OpenClaw is built on the OpenClaw open-source framework with a skill/plugin system, starting at $15.8/mo. Hermes is powered by Nous Research, lightweight and budget-friendly, starting at $4.2/mo. Both support 8+ messaging platforms.

Which messaging platforms are supported?

Telegram, Discord, WhatsApp, WeChat, Feishu, WeCom, DingTalk, and QQ.

Is there a free trial?

Yes. Both OpenClaw and Hermes include a 7-day free trial.

Simple, Transparent Pricing

Per-assistant billing. Choose what you need. No hidden fees.

Launch your AI agent today

Start chatting with your own AI agent today. Free trial available — no credit card needed to explore.