Manus is a general-purpose AI agent. You describe a goal and the agent plans the steps and carries them out in a cloud environment, so it can browse, write and run code and produce files while you do other work. The product menu lists web apps, AI slides, image, music and video tools, a video editor, a browser operator called My Browser, Wide Research for larger research jobs, Mail Manus, a Slack integration and automations. It is available as a web app, a desktop app and a mobile app, and the company introduced Manus 2.0 as its current agent generation.
Work is paid for with credits. The help center says a task's cost depends on the language-model tokens, virtual machine time and third-party APIs it uses, that credits are only consumed while a task is actively running, and that credits are refunded when a task fails because of a technical problem on Manus's side. The Free plan includes Chat Mode and the Manus 2.0 Lite agent, 300 refresh credits per day, one concurrent task and two scheduled automations. Paid plans start at $20 per month for 4,000 monthly credits plus the daily refresh credits, with higher tiers at $40 and $200 per month; paid plans add Manus 2.0 and 2.0 Max, 20 concurrent tasks and 20 scheduled tasks. Annual billing is listed at 17 percent less, and a Team plan adds SSO, usage analytics and an opt-out from data training.
Meta announced an agreement to buy Manus in December 2025, according to press reports, while the site footer still names Manus AI. The pricing page shows SOC 2 Type II and ISO 27001 badges. Because long agent runs consume credits at a variable rate, it is worth checking a task's likely cost before starting a large job.
Yes, with limits. The Free plan gives 300 refresh credits per day, Chat Mode and the Manus 2.0 Lite agent, one concurrent task and two scheduled automations. Paid plans start at $20 per month for 4,000 monthly credits.
The pricing page does not state a maximum run time. Tasks draw credits while they are active, so the practical limit is your credit balance and the plan's concurrency limit of one task on Free or 20 on paid plans.
Press reports in December 2025 said Meta agreed to acquire Manus. The website footer still reads Manus AI, so check the current terms and privacy policy before sending sensitive data.
Website: https://manus.im/
Company: Manus AI
Cloud agent for long research, website, slide and media tasks with scheduling. Free plan runs the Lite agent on 300 daily credits; costs vary per task. Paid from $20/month for 4,000 credits.
| Pricing model | Freemium |
|---|---|
| Free tier | Yes · 300 refresh credits per day on Manus 2.0 Lite; 1 concurrent task, 2 scheduled automations |
| Starting price | $20 / month |
| Data residency | Not verified |
| Compliance (as stated by vendor) | SOC 2 Type II (claimed), ISO 27001 (claimed) |
| Integrations | Slack |
| Public API | Yes |
| Self-hostable | Not verified |
| Official site | Not verified |
| Can a general AI assistant replace it? | A chat assistant can draft one answer at a time. Manus runs multi-step jobs in a cloud machine and returns files, which a plain chat session cannot do on its own. |
| Last verified | 2026-10-12 by editor |
Genspark is an AI workspace with a Super Agent that plans and completes multi-step tasks, plus AI slides, sheets, docs and meeting notes. The Free plan gives 100 credits a day up to a lifetime limit; Plus is $24.99 a month.
Visual automation platform for building workflows and AI agents by drag and drop, prompt or code, with more than 3,000 prebuilt app connections. The Free plan includes 1,000 credits a month.
No-code automation platform that connects more than 9,000 apps with multi-step workflows called Zaps, plus AI agents, tables, forms and chatbots. The Free plan includes 100 tasks a month.
Workflow and AI agent builder that you can run in n8n's cloud or on your own servers, with a visual canvas and code steps. A self-hosted Community Edition is free; cloud plans start at 20 euros a month.
A current review of AI Agent Store as a directory, hosted-agent platform, starter-kit catalog, agency marketplace, and agent-task service, with security and procurement checks.
Open Interpreter is an open-source terminal agent, forked from Codex, that reads and edits files and runs commands inside an OS-level sandbox with approvals. It is free under Apache 2.0 and works with local models through Ollama or LM Studio.
Need the outcome rather than the software?