Manus AI Agent Current Status (May 2026): Live Product, Unsettled Ownership

Manus AI sits in an unusual position in May 2026. The product is live, subscriptions are selling, new features shipped in the last month, and the public site at manus.im presents a full commercial surface. At the same time, the company’s ownership story is messier than it was six months ago, and that mess has real implications for anyone evaluating whether to build workflows around it.

This article is a status report. It covers where Manus stands right now: what the product does, what it costs, what changed recently, what the ownership situation looks like, and what all of that means if you are deciding whether to use it. It is deliberately separate from our Manus AI Agent Review (which covers hands-on testing and reliability) and our Manus AI Agent Capabilities guide (which catalogs every feature in detail). Think of this as the decision-context layer.

The Short Version

If you need the answer in under thirty seconds:

• Product status: Live and actively maintained at manus.im. All plans, features, and integrations are operational.
• Ownership status: Unresolved. Manus announced it was joining Meta in December 2025. Chinese regulators reportedly challenged the deal in late April 2026. The outcome is not settled.
• Capability status: Substantially expanded since launch. Wide Research, Browser Operator, a public API, team plans, and several integrations have all shipped in 2026.
• Risk assessment: The product works well today. The ownership question introduces medium-term continuity risk. Short-term use is straightforward; long-term dependency requires contingency planning.

Product Availability: What You Can Actually Use Right Now

Manus operates four subscription tiers as of May 2026:

Plan	Price	Credits	Best For
Free	$0	300/day	Trying the agent, simple tasks
Starter	$39/month	3,900/month	Regular research and coding tasks
Pro	$199/month	19,900/month	Heavy automation, API access
Scale	$399/month	39,900/month	Teams, high-volume workflows

All plans are available at manus.im with immediate signup. There is no waitlist. The free tier is generous enough to run real tasks — you get 300 credits per day, which covers several autonomous agent runs on straightforward prompts.

The product surface has not been stripped back or altered since the Meta announcement. If anything, the pace of feature releases accelerated in early 2026, which suggests the team continued shipping through the transition period.

What Has Changed in 2026

Manus entered 2026 as a capable autonomous coding agent. It has since expanded into a broader automation platform. Here is what shipped or materially changed:

Wide Research

Wide Research is the standout addition. Instead of sending a single agent to research a topic, Manus deploys hundreds of parallel sub-agents, each running in its own virtual machine with independent internet access. The result is that you can ask Manus to analyze 500 competitors, compare 200 product specs, or synthesize research across dozens of sources, and it handles the parallelization automatically.

This directly addresses the context window limitation that constrains every major AI assistant. ChatGPT, Claude, and Gemini all hit quality degradation when you ask them to process hundreds of items in a single session. Manus sidesteps the problem by distributing work across independent agents and merging the results.

Wide Research is available on Starter plans and above.

Browser Operator

Browser Operator is a Chrome extension that connects Manus to your active browser session. Instead of operating in a sandboxed environment, Manus can see and interact with tabs you already have open — using your local IP address and your existing login sessions.

This is useful for tasks like extracting data from logged-in dashboards, filling out forms on authenticated sites, or automating workflows across web apps that do not have public APIs. You grant permission before each session, and the agent operates within the tabs you specify. It does not store credentials independently.

Browser Operator is available on Starter plans and above.

Public API

Manus opened a public API at open.manus.ai/docs. It supports programmatic task submission, status polling, and result retrieval. This moves Manus from a tool you use interactively to one you can embed in automated pipelines.

API access requires a Pro plan or above.

Team Plans and Enterprise Features

Manus added team plans with shared credit pools, single sign-on, and collaboration features. This is a signals-maturity move: individual developer tools do not usually bother with SSO and shared billing unless they are seeing adoption from teams and organizations.

Team plans are available on custom pricing.

Integrations: Mail Manus and Slack

Mail Manus lets you send tasks to your Manus agent by email. Slack integration lets you assign tasks from Slack messages. Both are available on all plans.

These are smaller features individually, but they matter for the workflow question. If you can delegate tasks to Manus without opening a browser tab, the friction of using it drops substantially. The difference between “I have to go to manus.im and type a prompt” and “I forward this email and get results” is the difference between occasional use and habitual use.

The Ownership Situation: What We Actually Know

This is the part that most status updates either gloss over or over-dramatize. Here is what the public record shows, with sources:

1. December 29, 2025: Manus announced it was joining Meta. The announcement said Manus would continue operating its subscription service through its app and website.

2. January through March 2026: Manus continued shipping features, updating pricing pages, and maintaining the public product. No regulatory action was reported during this period.

3. April 27, 2026: Axios reported that Chinese regulators ordered Meta to unwind the Manus acquisition. The report cited regulatory concerns that had not been previously public. Meta responded that the transaction complied with applicable law and that it expected an appropriate resolution.

4. May 2026 (current): The product remains live. No shutdown, redirect, or feature removal has occurred. The ownership dispute between Meta and Chinese regulators has not produced a public resolution.

What This Means Practically

The ownership question and the product question are separate, and conflating them leads to bad decisions.

For the product: Manus works. It is maintained. Features are being added. The pricing has not changed. If you need an autonomous AI agent for a project this month or this quarter, the product does the job.

For the ownership: The Meta acquisition is contested. If Chinese regulators succeed in unwinding the deal, Manus could end up independent again, sold to a different buyer, or facing structural changes. None of these outcomes necessarily means the product disappears — but any of them could mean pricing changes, feature prioritization shifts, or organizational disruption.

The decision framework:

• If you are evaluating Manus for a short-term project (under 3 months), the ownership risk is low. Use the product on its merits.
• If you are building ongoing workflows that depend on Manus, build in an alternative path. If Manus changes pricing or discontinues a feature, you need a way to switch. This is just vendor management hygiene — you should do it for any single-vendor dependency, not just Manus.
• If you are making a team or enterprise purchase with a 12-month commitment, factor the ownership uncertainty into your vendor risk assessment. This does not mean “do not buy.” It means “document the contingency.”

How Manus Compares to Alternatives Right Now

The autonomous AI agent space has matured fast. Manus is not the only option, and whether it is the right one depends on what you are trying to do.

Manus vs. ChatGPT

ChatGPT is a conversational assistant. Manus is an autonomous agent. The difference matters. ChatGPT answers questions, generates text, and can write code snippets. Manus plans multi-step workflows, executes them independently, browses the web, runs code in virtual machines, and deploys applications.

If you need a research assistant that answers questions accurately, ChatGPT is faster and cheaper. If you need something that takes a goal and runs with it without constant prompting, Manus is the better fit.

Manus vs. Bolt and Lovable

Bolt and Lovable are focused on code generation for web applications. They are faster and more polished for that specific use case. Manus covers a broader scope — research, design, browser automation, presentations — but is less specialized for pure web app generation.

If you are building a web app from a design and need clean code fast, Bolt or Lovable are probably the better choice. If you need an agent that can handle research, analysis, code, and deployment as part of a single workflow, Manus is more capable.

Manus vs. Replit Agent

Replit Agent is the closest direct competitor. Both operate as autonomous coding agents that can build and deploy applications. Replit has the advantage of being built on top of a mature cloud development environment, which gives it better infrastructure for hosting and iteration. Manus has the advantage of broader non-coding capabilities like Wide Research and Browser Operator.

For pure coding projects, Replit Agent may have an edge in reliability. For mixed workflows that combine research, analysis, and code, Manus covers more ground.

Capabilities Snapshot

For readers who want the feature summary without reading the full capabilities guide:

Capability	Status (May 2026)	Plan Required
Autonomous coding	Live	All plans
Wide Research (parallel agents)	Live	Starter+
Browser Operator (Chrome extension)	Live	Starter+
AI Design (UI mockups, design assets)	Live	All plans
Presentations (Nano Banana Pro)	Live	All plans
Mail Manus (email task delegation)	Live	All plans
Slack integration	Live	All plans
Public API	Live	Pro+
Team plans with SSO	Live	Custom

Known Limitations

Status reports are only useful if they are honest about what is not working. Based on our testing and public reports:

Coding reliability is inconsistent for complex projects. Manus handles standard CRUD applications and prototyping well. It struggles with complex state management, multi-service architectures, and edge cases in production-grade code. Our hands-on review has specific test results.

Credit consumption can be unpredictable. Autonomous agents run until they decide they are done. A task that looks simple can consume more credits than expected if the agent takes an indirect path. This is less of a problem on the Scale plan and more of a problem on the Free or Starter tiers.

Browser Operator has scope limits. It works well for straightforward tasks in authenticated sessions, but it can get confused by complex multi-page workflows, CAPTCHAs, or sites with aggressive bot detection.

The API is relatively new. Documentation is solid, but the API does not yet expose all the capabilities available through the web interface. If you are building production integrations, verify that the specific features you need are available through the API endpoint.

Ownership uncertainty is itself a limitation. Even if the product continues to work perfectly, the unresolved acquisition creates planning uncertainty that would not exist with a comparably capable tool from a stable, independent company.

What to Watch Over the Next 60 Days

If you are tracking the Manus situation, here are the signals that would change the status assessment:

1. A regulatory resolution — either Chinese regulators approve the Meta deal, or it is formally unwound. Either outcome reduces uncertainty.
2. A change in product availability — if manus.im stops selling subscriptions, removes features, or redirects to a Meta domain, that would signal a material shift.
3. Pricing changes — the current pricing has been stable through 2026. A sudden price increase or plan restructuring could indicate cost pressure from the ownership dispute.
4. Feature velocity changes — Manus shipped Wide Research, Browser Operator, and the API in quick succession. If feature releases slow or stop, that could signal team disruption.
5. Competitive moves — if OpenAI, Anthropic, or Google ship comparable autonomous agent features that match Manus’s breadth, the value proposition shifts independently of what happens with the Meta deal.

Summary

Manus AI in May 2026 is a capable, live, actively maintained autonomous AI agent platform. It does more than it did at launch, the pricing is transparent, and you can start using it today without a waitlist.

The unresolved ownership situation is real and should be part of your evaluation, but it should not be the only thing you evaluate. The product works. The question is whether you are comfortable with the medium-term continuity risk, and that answer depends on your timeline and your tolerance for vendor uncertainty.

For detailed feature breakdowns, see our Manus AI Agent Capabilities 2026 guide. For hands-on project testing and reliability scores, see our Manus AI Agent Review 2026.

Get the action plan for Manus Ai Agent Current Status 2026

Get the exact implementation notes for this topic, plus weekly briefs with cost-saving workflows.

Send me the action plan