Manus Review 2026: Pricing, Features and Verdict
The general purpose AI agent, judged on its help center pricing, docs, app store and Trustpilot ratings, and reporting on its ownership.
By , research by the BestBots team ยท Updated Oct 18, 2026
Fact checked against each company's published pricing, documentation and verified user ratings. How we score
Manus is the best AI agent for big, parallel jobs in October 2026. Wide Research splits a task across many agents, paid plans run 20 tasks at once, and Manus 2.0 builds sites, slides, videos and games from a prompt. Pro starts at $20 per month for 4,000 credits. The downsides: credit burn you cannot predict, poor Trustpilot scores and a year of ownership upheaval.
Manus is the strongest pick for big parallel jobs: Wide Research splits work across many agents, and paid plans run 20 tasks at once from $20 per month. The tradeoffs are unpredictable credit burn, a 1.2 Trustpilot score from 218 reviews driven by billing and support complaints, and ownership churn after Meta's acquisition was unwound in 2026.
No user reviews yet. Write a review
- Starting price
- $20 per month
- Pricing model
- per credit
- Free plan
- Yes
- Free trial
- 7 days on the $40 Pro tier
- Price
- Free with 300 daily credits; Pro from $20 per month for 4,000 credits
- Platforms
- Web, iOS, Android, desktop app (Manus Studio)
- Agent mode
- Agent Mode with Cloud Browser, Browser Operator in your own browser, Wide Research
- Built on
- Manus 2.0, 2.0 Max and 2.0 Lite agents
- Privacy controls
- Training opt out on Team; Browser Operator needs per session authorization
- Availability
- Independent again after the Meta deal was unwound (announced August 11, 2026)
What it is
Manus is a general purpose AI agent from Butterfly Effect Pte. Ltd., a company founded in China and based in Singapore. You give it a goal and it plans and executes in its own cloud environment: browsing, writing code, building documents and deploying websites, with a live view of what it is doing.
It has two modes. Chat Mode answers quick questions. Agent Mode does the work, running one of three agents: Manus 2.0 Lite (the only one on Free), Manus 2.0 and Manus 2.0 Max.
Ownership. Meta acquired Manus on December 29, 2025. China's National Development and Reform Commission ordered the deal withdrawn in April 2026, CNBC reported, and on August 11, 2026 Manus said it would "soon return to operating as an independent company." As part of the separation, data generated by some users on or after December 29, 2025 was deleted between August 23 and 24, 2026, after a backup window. If you are evaluating Manus for a business, factor in that history.
Pricing explained
Manus bills in credits. Every plan includes Chat Mode.
| Plan | Price | Credits | Limits |
|---|---|---|---|
| Free | $0 | 300 per day | 1 concurrent task, 2 scheduled automations, Manus 2.0 Lite only |
| Pro | From $20 per month | 4,000 per month | 20 concurrent tasks, 20 scheduled automations, all three agents |
| Pro | From $40 per month (7 day free trial) | 8,000 per month | Same as above |
| Team | From $20 per seat per month | Shared pool | Pro features plus SSO, training opt out, usage analytics, access control |
Annual billing saves 17% on Pro and Team. Higher credit tiers exist above 8,000 per month; Manus's help center says the live pricing page governs. Older Basic and Plus plans were renamed Pro, and existing members renew at their old prices.
How credits work. Credits are consumed by task complexity and duration. Plan credits reset monthly; add on packs never expire. The docs say the dashboard estimates usage before you start a task. Phone usage through Manus's Cue app is billed separately from credits.
The real cost. Both tiers work out to 200 credits per dollar, so $20 buys 4,000 credits. What a credit buys is the hard part: a short research question might cost little, while a website build with several fix passes can use thousands. Several Trustpilot reviewers describe spending thousands of credits on a single troubled build. Budget a buffer, start on the cheapest paid tier and watch your usage history before upgrading. See our Manus pricing guide for worked examples.
Key features
Wide Research
Manus's signature feature. Instead of one agent working through a long list and losing quality as its context fills, Wide Research gives each item its own agent and context, runs them in parallel and assembles the results into a table or report. Manus's examples include a database of 250 AI researchers and a comparison of 100 sneaker models.
Cloud Browser
A browser that runs in Manus's cloud and operates like a person: visiting sites, clicking, filling forms and extracting data. You can log into your own accounts in it so Manus can act for you, and when it hits a CAPTCHA, SMS code or multi factor prompt it asks you to take over.
Browser Operator
An extension that lets Manus work inside your local browser with your existing logins, in a dedicated tab. Manus recommends it for authenticated sites like LinkedIn or a CRM, and the isolated Cloud Browser for general research. You authorize every session and can stop the agent by closing the tab; actions are logged.
Manus Studio and Manus 2.0
The desktop app is now Manus Studio, a shared workspace for documents, spreadsheets, PDFs, slides, websites, code, games and video. Manus 2.0 added a Video Editor with an editable timeline, a Game Dev environment with multiplayer hosting on a purchasable Cloud Computer, and Automations that start when an email arrives, ad performance changes, a calendar event fires, or a Slack or Notion update lands. Manus says its new Cascade architecture used 23.2% fewer tokens and cost 32% less to run in one tested configuration.
Scheduled tasks and concurrency
Paid plans run 20 tasks at once and 20 scheduled automations. This is the number that sets Manus apart from agents that queue one job at a time.
What users say
- App Store: 4.7 from 39,344 ratings for the Manus mobile app.
- Trustpilot: 1.2 from 218 reviews.
The gap is stark. Mobile users rate the app experience highly. Trustpilot reviewers, often paying customers with billing problems, describe credits consumed by failed or looping tasks, refund disputes, projects flagged for policy issues and slow escalation to human support. Recurring praise across sources: the output quality on research and slides, Wide Research, and parallel tasks. Recurring complaints: credit unpredictability and support.
Privacy and security
Training. Manus lists "Data Training Opt-Out" as a Team plan feature. We did not find a per user training opt out documented for Free or Pro on the pages we read, so assume your data may be used unless you are on Team.
Your accounts. Cloud Browser logins and Browser Operator sessions give Manus the same access you have. Browser Operator requires authorization for every session. Use it only on sites the task needs.
Data continuity. The August 2026 deletion of some users' data during the Meta separation is a reminder to keep your own copies of anything important you build in Manus.
Prompt injection. Manus reads web pages and acts on them, so the same indirect prompt injection risk that Brave documented across agentic browsers in 2025 applies: hidden text on a page can try to redirect the agent. Give specific instructions, avoid open ended access to email or banking, and review results before anything is sent or published.
Who it is for
- Researchers and analysts who need the same deep work done on dozens or hundreds of items.
- Founders and marketers who want a website, slides or a short video from a prompt.
- Power users who run many tasks at once and will monitor credit use.
Who should skip it
- Anyone who needs a fixed monthly cost. Credit burn varies by task. ChatGPT agent on Plus is a flat $20.
- Companies that need stable vendor ownership and published certifications. We did not find a SOC 2 report on the pages we read.
- People who rely on fast human support. Trustpilot reviews flag it as a weak spot.
Alternatives
- ChatGPT agent: ChatGPT Work for $20 per month on Plus with 1,400+ plugins.
- Genspark: Super Agent plus a weekly Free Quota that does not spend credits.
- Perplexity: Computer agent with cited research.
- Lindy: an AI assistant for inbox, calendar and CRM work.
See all Manus alternatives, or compare ChatGPT agent vs Manus and Manus vs Genspark.
Verdict
Manus is the most capable agent for wide, parallel work and creative builds, and its free plan refreshes 300 credits every day so you can judge it before paying. Credit pricing makes the real cost hard to predict, Trustpilot users report billing and support pain, and the company just unwound a $2 billion acquisition. Use it for research and builds where its parallelism pays off, keep copies of your work, and check usage history before moving up a tier.
How Manus scores
- Value
- 3.8
- Features
- 4.5
- Ease of setup
- 4.2
- Integrations
- 4.0
- Trust
- 3.1
Pros and cons
Pros
- Wide Research deploys many agents in parallel, each with its own context
- 20 concurrent tasks and 20 scheduled automations on paid plans
- Choice of a sandboxed Cloud Browser or the Browser Operator extension in your own browser
- Free plan refreshes 300 credits every day
Cons
- Credit use depends on task complexity, so costs are hard to predict
- Trustpilot score of 1.2 from 218 reviews, with complaints about credit burn, refunds and support
- Training opt out is listed only on the Team plan
- Some users' data was deleted in August 2026 as part of the Meta unwind
Manus plans and pricing
| Plan | Price | What you get |
|---|---|---|
| Free | $0 | 300 credits per day, 1 concurrent task, 2 scheduled automations, Chat Mode and Manus 2.0 Lite |
| Pro | From $20 per month | 4,000 credits per month, Manus 2.0 Max, 2.0 and 2.0 Lite, 20 concurrent tasks, 20 scheduled automations, Wide Research |
| Pro (8,000 credits) | From $40 per month | 8,000 credits per month, 7 day free trial, same features |
| Team | From $20 per seat per month | All Pro features, SSO, data training opt out, usage analytics, shared credit pool |
What users say
Manus reviews from users
No user reviews yet
Be the first to review ManusManus alternatives
ChatGPT agent4.5
Gemini in Chrome4.4
Brave Leo4.2
Comet4.2
Copilot Mode in Edge4.2
Genspark4.1Frequently asked questions
Is Manus AI free?
There is a free plan with 300 credits refreshed daily, one concurrent task, two scheduled automations, Chat Mode and Manus 2.0 Lite in Agent Mode. Paid Pro plans start at $20 per month.
How much does Manus cost?
Pro starts at $20 per month for 4,000 credits or $40 per month for 8,000 credits, with higher credit tiers available and 17% off for annual billing. Team starts at $20 per seat per month. The $40 tier has a 7 day free trial.
How many credits does a Manus task use?
Manus says it depends on task complexity and duration, so there is no fixed number. Its docs say the dashboard gives estimates before you start, and recommend specific prompts, Chat mode for simple questions and batching similar tasks to save credits.
Do Manus credits roll over?
Plan credits reset every month. Add on credit packs you buy never expire, according to Manus's docs.
Is Manus owned by Meta?
Not anymore. Meta acquired Manus on December 29, 2025. China's National Development and Reform Commission told the parties to withdraw the deal in April 2026, CNBC reported, and Manus announced on August 11, 2026 that it would return to operating as an independent company.
Is Manus safe to use?
Treat it like any agent with access to your accounts. The Browser Operator extension uses your logged in sessions, so authorize only the sessions a task needs and watch its tab. Training opt out is listed on the Team plan, and some users had data deleted in August 2026 as part of the Meta unwind.
What is Manus Wide Research?
Wide Research splits a job with many similar items, such as profiling 250 people or comparing 100 products, across many agents running in parallel, each with its own context, so item 250 gets the same depth as item 1.
Why are Manus Trustpilot reviews so bad?
As of October 2026 Manus scores 1.2 on Trustpilot from 218 reviews. Recent reviews describe credits consumed by failed or looping tasks, disputes over refunds and slow human support. App Store users rate the mobile app much higher at 4.7 from 39,344 ratings.