Manus

Manus is a general-purpose AI agent: give it a goal rather than a single question and it plans the job, browses the web, runs code, opens files and delivers a finished result — a researched report, a spreadsheet, a slide deck or a small website — while you do something else. It is built for delegated, multi-step work, not one-shot chat answers.

Key facts: Manus launched invite-only in early 2025 and moved to credit-based paid tiers in May 2025, quickly becoming one of the most talked-about agent products of the year. Along the way it added Plan Mode for editing the agent's plan before it runs, Wide Research that splits large research jobs across many sub-agents in parallel, scheduled tasks, and a desktop mode that can work on your own machine.

Plans: Free costs $0 with 300 credits refreshed daily, enough for small tasks. Pro starts at $20/month for 4,000 monthly credits, with larger tiers at $40/month for 8,000 credits and $200/month for 40,000 credits, and Team plans from $20/seat/month with shared pools and admin controls. Every task spends credits based on its length and the tools it uses.

Best for: people who want to hand off genuinely multi-step jobs — research, comparisons, data gathering, content production — and check the result rather than do each step. Watch out: credit burn is hard to predict, with a single deep-research run reported to consume several hundred credits, so start on the free tier and learn your real usage before committing to a paid tier.

Pros

• A general-purpose agent: give it a goal and it plans the job, browses the web, runs code, opens files and delivers a finished result such as a researched report, spreadsheet, slide deck or small website.
• Plan Mode lets you edit the agent's plan before it runs.
• Wide Research splits large research jobs across many sub-agents in parallel.
• Scheduled tasks and a desktop mode that can work on your own machine extend it beyond chat.

Watch out for

• Credit burn is hard to predict, with a single deep-research run reported to consume several hundred credits.
• Start on the free tier and learn your real usage before committing to a paid tier.
• Agent output still needs checking before you act on it or publish it.

Best for

People who want to hand off genuinely multi-step jobs, such as research, comparisons, data gathering and content production, and check the result rather than do each step. Who should skip it: users who only need one-shot chat answers.

Pricing

Free costs $0 with 300 credits refreshed daily, enough for small tasks. Pro starts at $20/month for 4,000 monthly credits, with larger tiers at $40/month for 8,000 credits and $200/month for 40,000 credits, and Team plans from $20/seat/month with shared pools and admin controls. Every task spends credits based on its length and the tools it uses. Check the official pricing page for current pricing.

FAQs

What is Manus? A general-purpose AI agent that takes a goal, plans the job, browses, runs code and delivers a finished result while you do something else.

What is Plan Mode? A mode for editing the agent's plan before it runs.

What is Wide Research? A feature that splits large research jobs across many sub-agents in parallel.

How do credits work? Every task spends credits based on its length and the tools it uses, so start on the free tier and learn your real usage first.