Can OpenClaw Run Multiple Agents?

OpenClaw FAQ

Yes! OpenClaw supports multiple agents with different models, skills, and configurations. Run a coding agent, a personal assistant, and a home automation agent simultaneously.

Multi-Agent Architecture

OpenClaw supports running multiple agents, each with their own:

  • AI model (e.g., Opus for complex tasks, Haiku for simple ones)
  • Skill set
  • Memory workspace
  • Channel assignments

Sub-Agents

Agents can spawn sub-agents for specialized tasks. A main agent might delegate coding to a coding sub-agent and research to a research sub-agent.

Use Cases

  • Different agents for different messaging channels
  • Specialized agents for coding, research, home automation
  • Team deployments with per-member agents
// Multi-agent config
{
  "agents": {
    "defaults": {
      "model": { "primary": "anthropic/claude-sonnet-4" }
    },
    "coder": {
      "model": { "primary": "anthropic/claude-opus-4-6" },
      "workspace": "~/.openclaw/workspace-coder"
    },
    "assistant": {
      "model": { "primary": "anthropic/claude-haiku-3.5" }
    }
  }
}

Try OpenClaw — free and open source

Sign up and launch your AI agent in under a minute. No credit card required.

Get Started Free →

Related