The best Meta Muse alternatives outside the US are ChatGPT Work, Google's Gemini Spark, Claude, Grok Bot, Perplexity Teammates, Lindy, Manus and the open-source agents OpenClaw and Hermes, because all of them can be used in places like India, the UK and Europe where Muse is not available yet. If you want more than one assistant, Vestra is the alternative that gives you a whole team of agents, led by a chief, from $19 a month and billed in rupees with GST in India.
Meta Muse launched on 8 September 2026. Today Meta says it is available to adults in the US and Canada, and since 27 October in Singapore. Everyone else is on the outside looking in. This guide is for them: what Muse is, where it works, when it might reach India, and which alternatives you can use today instead.
There is one thing worth saying before the list. Muse is a personal agent. It is built for one person and it works alone. Almost every alternative on this page is the same kind of thing with a different badge: ChatGPT Work, Gemini Spark, Claude Code, Codex, Grok Bot, OpenAI Dots and Hermes each serve the one person who started them. So the real choice is not only "which personal agent can I get here?" It is also "do I want one assistant, or a team that works together?"
Quick answer
What are the best Meta Muse alternatives outside the US? Pick by what you need done:
- For a whole team of agents that work together: Vestra. A chief you name runs teammates that each hold a whole job, on Vestra's own engine, Claude Code or Codex. From $19 a month, billed in rupees with GST in India.
- For one personal agent inside the chat app you already use: ChatGPT Work. Limited Work comes even on the free plan and on Go, and full Work starts with Plus at $20 a month.
- For a background agent in Gmail, Docs and Sheets: Gemini Spark, for Google AI Pro and Ultra subscribers, rolled out in India from late July 2026.
- For coding and long tasks: Claude, with Claude Code on every paid plan, supported in India and Canada.
- For agents you brief one by one: Grok Bot (with an X subscription from $30 a month) or Perplexity Teammates (Perplexity Max, $200 a month).
- For one assistant with its own phone line: Lindy, from $29.99 a month per active user, with a US or Canadian number.
- For full control and no subscription: OpenClaw or Hermes Agent, both free and open source, which you host and secure yourself.
The short version: if you only want "Muse, but available here", ChatGPT Work and Gemini Spark are the closest. If you want agents that work together instead of alone, Vestra is built for that.
Key takeaways
- Muse works in very few places. Meta's own pages say it is available in the US and Canada for people 18 and over, and Meta launched it in Singapore on 27 October 2026. It is not available in India, the UK, the EU or Australia yet.
- India is on a waitlist. Indian press reports say Meta has listed rupee prices for Muse, but Meta has not confirmed a launch date or those prices. A Meta India leader has said India is a priority for next year, according to a report.
- Most alternatives are personal agents too. ChatGPT Work, Gemini Spark, Claude Code, Codex, Grok Bot, OpenAI Dots and Hermes each work for one person. Their memory, history and bill belong to that one person.
- Availability is only the first filter. After "can I use it here?", the questions that matter are: does it work in my tools, does it ask before it spends, can I see what it did, and can it work with other agents?
- The open-source agents work anywhere, at a cost. OpenClaw and Hermes run wherever you put them, but you run the server, the keys and the security.
- A team beats a pile of assistants. Five personal agents means five briefs, five logins and five bills, and you carry work between them. A chief that runs the others turns that pile into a team.
- Vestra is the one office where agents on different engines work as teammates. Your chief runs teammates on Vestra's own engine, Claude Code or Codex, under one set of limits and approvals, with a receipt for every piece of work. Plans start at $19 a month, and in India they are billed in rupees with GST included.
Contents
- What are the best Meta Muse alternatives outside the US?
- What is Meta Muse?
- Where is Meta Muse available today?
- Is Meta Muse available in India?
- When will Meta Muse launch in India, and what will it cost?
- Why does a new AI agent launch in only a few countries?
- Can you use Meta Muse from outside the US anyway?
- What should you look for in a Meta Muse alternative?
- What is the difference between a personal agent and an AI team?
- How do the 11 Meta Muse alternatives compare?
- Which Meta Muse alternatives can you use outside the US?
- Which Meta Muse alternative is best for a small business in India?
- Which alternative is best if your business runs on WhatsApp?
- Which alternative is best for developers?
- Which Meta Muse alternatives are free or cheapest?
- How do you pay for an AI agent in rupees?
- How do you run several AI agents together instead of one at a time?
- What does a week look like with one personal agent, and with a team?
- How do you switch from Meta Muse, or stop waiting for it?
- What is live in Vestra today, and what is still being built?
- Frequently asked questions
- Glossary
What are the best Meta Muse alternatives outside the US?
The best Meta Muse alternatives outside the US are the agents you can sign up for today in your country. For one personal agent, that means ChatGPT Work, Gemini Spark, Claude, Grok Bot, Perplexity Teammates, Lindy or Manus. For self-hosting, OpenClaw and Hermes. For a team of agents that work together, Vestra.
Here is the full list of 11, in the order this guide covers them:
- Vestra: a chief that runs a team of AI teammates, each on the engine that suits its job.
- ChatGPT Work: OpenAI's mode in ChatGPT for handing over one task at a time.
- OpenAI Dots: an always-on agent that comes with ChatGPT Pro.
- Gemini Spark: Google's 24/7 background agent in Gmail, Docs and Sheets.
- Claude: Anthropic's assistant, with Claude Code and scheduled tasks on paid plans.
- Grok Bot: xAI's Bots that you create and brief one by one.
- Perplexity Teammates: agents that own ongoing work, in research preview on Perplexity Max.
- Lindy: one AI employee for each person's inbox and calendar, with a phone line.
- Manus: a general-purpose agent with Free, Pro and Team plans.
- OpenClaw: a free, open-source assistant you host yourself.
- Hermes Agent: Nous Research's free, open-source agent that runs wherever you put it.
Each one gets its own section below, with what it is, where it works, what it costs and the catch. Every competitor fact comes from that company's own pages, checked on 14 November 2026, unless it says otherwise.
Two warnings before you read on. First, "available in your country" is not one fact. A product can sign you up but hold back features, like Muse's calls to businesses, which Meta has opened only to paid users in the US. Second, prices change often. Treat every price here as a snapshot and check the company's own page before you pay.
What is Meta Muse?
Meta Muse is a personal AI agent from Meta, launched on 8 September 2026. It acts for you: it can send emails, fill in forms, shop and plan trips, using its own secure cloud computer. Meta calls it the world's first personal AI agent built for everyone. It runs on Muse Spark, Meta's own model.
A personal AI agent is software that does tasks for one person, not just answers questions. You give it a goal, it works out the steps and it takes actions in apps and websites. Muse fits that definition well.
What Meta says Muse does, from its own launch post and pages:
- Acts on your behalf. It works in your connected email and calendar, fills in forms and shops for you.
- Has its own computer. Muse works inside Muse Secure VM, a dedicated virtual machine that holds both the agent and your data. It keeps working after you close the app.
- Asks before sensitive actions. Muse checks with you before it sends an email or makes a purchase. A separate agent Meta calls the Sentinel checks outbound activity, and Meta says nothing Muse does reaches the internet unless the Sentinel approves it.
- Lives where you already are. You reach it in the Muse app on iOS and Android, on the web at muse.ai and by messaging it in WhatsApp. Meta says it is coming soon to AI glasses.
- Remembers you. Muse keeps a memory you can edit and delete, in a file Meta calls Memory.md, and you can import memories from another assistant.
- Works on a schedule. It runs recurring tasks and can send you messages and ideas without being asked.
- Pays carefully. When it shops, Muse can make a one-time card number at checkout, so your real card number is not seen.
On 29 September 2026 Meta added Muse for Small Business, which it describes as a collection of skills and connectors for small business owners. Meta's page lists connectors to tools such as Shopify, QuickBooks, Stripe, Klaviyo, Canva, Notion and Slack, plus Facebook, Instagram and Meta ads. It helps with customer replies, product launches, content planning and cash flow, with approvals for actions like sending emails or making purchases and an audit trail of what the agent did.
On 5 November 2026 Meta added voice and expressive avatars in the Muse app, rolling out in the US first.
How much it costs: Meta says Muse is free for most of what people need, with subscription plans for people who want more. Meta's Help Center lists Power at $20 a month for 500 million Muse tokens a week and Maximum at $100 a month for 3 billion Muse tokens a week. That help page did not load for us on 14 November, so these prices come from our 9 November check; a 10 November Analytics India Magazine report gives the same US prices.
The thing to notice is the word "personal". Muse is built for one person. Muse for Small Business gives that one person skills and connectors for their business. It does not give them a team of agents that hand work to each other. We cover why that matters in the section on personal agents and teams.
For a side-by-side of Muse and Vestra on every feature, see Vestra vs Meta Muse.
Where is Meta Muse available today?
Meta Muse is available today in the US and Canada, for people aged 18 and over, and in Singapore, where Meta launched it on 27 October 2026. Some features are narrower still: voice and avatars started in the US, and calls to businesses are open only to paid users in the US. India, the UK, the EU and Australia are not on the list yet.
Here is what Meta's own pages say, checked on 14 November 2026:
- US and Canada. The Muse for Small Business page says Muse is "available today to people in the US and Canada who are 18 and over", and that Meta plans to expand to more markets without giving timing.
- Singapore. A Meta newsroom post dated 27 October 2026 says "Muse is available in Singapore from today", on iOS, Android and the web, with local connectors like Shopee, Grab, Singapore Airlines, Agoda, ShopBack, Klook, Trip.com and Lazada rolling out progressively.
- Voice and avatars. Meta's 5 November post says they are rolling out "in the Muse app in the US to start".
Note that Muse's small business page still lists only the US and Canada, even though Meta's newsroom says Muse launched in Singapore. Pages fall behind launches. If you are in a country near the edge of the list, check both.
A few features are narrower than the country list:
- Calls to businesses. Muse can now call a business for you, for example a cable company or a restaurant. A Business Upturn report from 14 November says Meta opened this to paying Muse users in the US on 12 November, and that it is not available in India. We could not find this on Meta's own pages, so treat it as reported.
- Local connectors. Each country gets its own partners. Singapore got Shopee and Grab. A business in Pune or Leeds would need its own local apps, which do not exist in Muse yet.
So for most of the world, the answer today is simple. Muse is not available to you, and Meta has not said when it will be.
Is Meta Muse available in India?
No. Meta Muse is not available in India as of 14 November 2026. Meta's own pages list the US, Canada and Singapore. Indian users are on a waitlist, according to Indian press reports, and Meta has not confirmed a launch date or Indian prices. Calls to businesses are not available in India either.
This is the question most people searching "meta muse india" want answered, so here is everything we found, and where it came from.
What Meta says. Meta's Muse for Small Business page says Muse is available to adults in the US and Canada and that Meta plans to expand to more markets, without giving timing. Meta's newsroom adds Singapore. Neither mentions India.
What the press reports say. These are not Meta's own words, so read them as reports:
- On 10 November 2026, Analytics India Magazine, citing The Hindu, reported that Meta has listed Muse subscription prices for India ahead of a launch: ₹1,799 a month for Power and ₹8,999 for Maximum, with subscriptions through Google Play and Apple's billing costing slightly more. The same report says Indian users are on a waitlist, that Muse's web and WhatsApp interfaces can be reached from India with some features restricted, and that Meta has not confirmed a launch date or the prices.
- On 14 November 2026, Business Upturn reported that Meta India head Arun Srinivas told Moneycontrol that bringing Muse to India would be a priority next year, but that the company is still weighing how to manage the rollout.
So the honest summary for India is: coming, probably, but not yet, at a price nobody has confirmed.
That leaves two practical choices for an Indian user or business. Wait, and use something else in the meantime. Or pick an alternative now and decide later whether Muse is worth switching to. The rest of this guide is about the second choice.
If you run a business in India, there is a third question worth asking. Even when Muse arrives, will a personal agent for one person be enough? We come back to this in the section on small businesses in India.
When will Meta Muse launch in India, and what will it cost?
Meta has not announced when Muse will launch in India or what it will cost there. A report quoting Meta's India head says India is a priority for next year. Indian press reports say Meta has listed Power at ₹1,799 a month and Maximum at ₹8,999, but Meta has not confirmed either the date or the prices.
Here is what we can say with care:
- Timing. "A priority next year", per the reported Moneycontrol interview. That is a statement of intent, not a date.
- Price. The reported ₹1,799 and ₹8,999 are close to the US prices of $20 and $100 a month at today's exchange rates. That may not be what Indian users see at launch. Meta may change them, add a cheaper tier or bundle Muse with something else.
- Features. Even after launch, expect some features to come later. Singapore got Muse with local connectors "rolling out progressively". Calls to businesses started in the US only. Voice started in the US only.
What this means for planning: if your work needs an agent this quarter, waiting for Muse is a plan with no date. It is fine to wait if you are curious. It is a risk if your business needs the help now.
One more point on cost. Muse prices its paid plans in Muse tokens, a weekly allowance. Tokens are a unit of model use, not a unit of work, so it is hard to know in advance what a task will cost. Compare that with tools that show a receipt per task. We cover how pricing models differ in the section on rupee billing.
Why does a new AI agent launch in only a few countries?
Meta has not said why Muse launched in only a few countries. In general, an agent that acts for you depends on local pieces: payment partners, local apps to connect, language support, data and consumer rules, and enough servers. Companies often launch where those pieces are ready and add countries one by one.
This section is general background, not Meta's explanation. But it helps you judge any agent, not just Muse.
An agent is different from a chatbot. A chatbot answers. An agent does things in the world: it books, buys, sends and signs up. Every one of those actions touches something local.
- Payments. An agent that shops needs a way to pay that works in your country. Muse's one-time card numbers, for example, depend on a payment partner. Card rules, UPI and local wallets differ from country to country.
- Local apps. Muse in Singapore launched with Shopee, Grab and Agoda. A Muse for India would need the apps people in India use every day. Connectors are built one partner at a time.
- Language. An agent has to understand how people in a country write and speak, including mixes of languages. Getting that right takes testing.
- Rules. Data, privacy and consumer protection rules vary. An agent that holds your email and acts in your name has a lot to get right in each place.
- Capacity. Agents run on expensive computers in the cloud. A company may hold back new countries until it has enough of them.
None of this is unique to Meta. OpenAI's own page says it introduced ChatGPT Go in India in August 2025 before launching it in the US, with prices "localized" in some markets. Google's Gemini Spark reached India in late July 2026 as an expansion to subscribers there. Every company is doing the same country-by-country work.
The practical lesson: when you pick an alternative, check two things, not one. Can you sign up from your country? And do the features you need, like calls, payments or a local app, work in your country too?
Can you use Meta Muse from outside the US anyway?
Not properly. Meta makes Muse available to people in the US, Canada and Singapore. Indian press reports say some Muse interfaces can be reached from India with features restricted, but Muse is not officially offered there. Getting around location limits can break a service's terms and put your account and data at risk, so we do not recommend it.
People ask this a lot, so here is a plain answer without the tricks.
What is reported. The 10 November Analytics India Magazine report says Muse's web and WhatsApp interfaces can be reached from India, with some features restricted, and that Indian users are on a waitlist. That is not the same as Muse being available to you.
Why workarounds are a bad idea for an agent. With a chatbot, the worst case is a lost conversation. With an agent, it is worse:
- It holds your logins. An agent you connect to Gmail, your calendar and your shop has the keys to a lot. If your account is limited or closed for breaking terms, untangling that is painful.
- It spends money. Payment features are tied to supported countries and partners. A purchase that fails halfway is worse than one that never started.
- Local features will not work. Connectors for US or Singapore apps do nothing for a business that sells on Indian or British platforms.
- Support is thin. If something goes wrong, a product that is not offered in your country is not obliged to help you.
The better path is to use an agent that is properly offered where you are, and to keep an eye on Muse for the day it launches. When it does, you can test it next to whatever you are using.
What should you look for in a Meta Muse alternative?
Look for six things: it is properly offered in your country, it works in the tools you use, it asks before it spends money or contacts someone new, it shows you what every task cost, it remembers what it learns, and it can work with other agents instead of alone. The last one is the one most people forget.
Here is a checklist you can use on any agent, including the ones on this page.
- Real availability where you are. Not "it loads", but "they offer it here". Check the supported-countries page, the payment options and which features are country-limited.
- Your tools, connected. Does it work in the email, calendar, CRM, books and chat apps you already use? If an app has no connector, can the agent use the website instead, on a computer of its own?
- Approvals you control. Does it ask before it spends money, deletes something or contacts someone new? Can you tune how much it does alone, task by task?
- Costs you can see. Can you cap spending per task and per day? Do you get a receipt that shows what each piece of work cost?
- Memory you can check. Does it remember what you taught it? Can you open that memory, see where each fact came from and fix it?
- Teamwork. Can it hand work to another agent and take work back? Can several agents share one memory, one set of rules and one bill? Or does each agent work alone for you?
Most personal agents score well on the first five. Muse itself asks before sending or buying, keeps an editable memory and makes one-time card numbers. The sixth is where personal agents fall down, because they were never designed to work with each other.
Two more questions that matter for businesses outside the US:
- Currency and tax. Can you pay in your own currency, with the right tax invoice? For an Indian business, an invoice in rupees with GST is easier for the books than a dollar charge on a card.
- Where it runs. If your data has to stay in a particular cloud or country, can the agent run there?
What is the difference between a personal agent and an AI team?
A personal agent works for one person and works alone: its memory, history and bill belong to that person, and it does not hand work to other agents. An AI team is several agents with separate jobs, run by one lead agent, sharing one memory and one set of rules. You talk to the lead, and it runs the rest.
This is the most important idea in this guide, so let us make it concrete.
A personal agent is like a very capable assistant who only ever talks to you. Muse is one. So are ChatGPT Work, Gemini Spark, Claude Code, Codex, Grok Bot, OpenAI Dots and Hermes. Each is excellent at what it does. And each one:
- knows only what you told it;
- keeps its own memory, in its own format, inside its own company's walls;
- has its own login to each of your apps;
- sends its own bill;
- cannot see what your other agents are doing, or hand them work.
An AI team is closer to a small company. There is a lead, which Vestra calls a chief, that knows your goals and hands out work. There are teammates, each hired for a whole job like research, sales follow-up or bookkeeping. They share one memory, so a lesson taught to one reaches all of them. They work under one set of rules and spend from one budget, with one record of what happened.
Here is the contrast at its starkest:
- Who you talk to. Personal agents: each one, separately. A team: your chief.
- Who carries work between agents. Personal agents: you, by copying and pasting. A team: the chief hands it over, with the context attached.
- Who checks the work. Personal agents: you. A team: another agent, a test, the source or a person, before it reaches you.
- What happens when you correct one. Personal agents: that one agent learns, maybe. A team: the lesson goes into the shared memory and every teammate follows it.
- How many logins, bills and rulebooks. Personal agents: one per agent. A team: one.
Vestra's handbook names the four jobs people end up doing when they run several personal agents: they brief each one with the same context by hand, carry work from one agent to another, chase every result to check it is done and right, and click through forms and settings for every request. That is the hidden cost of "I'll just use five assistants".
Some vendors have started adding teamwork inside their own walls. Claude Code has agent teams, which Anthropic's docs describe as experimental and off by default. Codex can spawn subagents inside one chat. Hermes has a Bot Mode with specialist Bots in group chats. These are real steps. But each one coordinates copies of the same product, for the one person who started it. None of them lets a Claude Code agent and a Codex agent share a memory, a budget and a boss.
That is the gap Vestra fills. In Vestra, a teammate's engine is the part that does the thinking, and you pick it teammate by teammate: Vestra's own, Claude Code or Codex are live today. One teammate can write code on Claude Code while another reviews it on Codex, both run by your chief, under one set of rules and one receipt. For a longer explanation, read What is an AI office? and What is an AI chief of staff?
So when you compare Muse alternatives, ask two questions. Which personal agent is best for me? And do I want a personal agent at all, or a team?
How do the 11 Meta Muse alternatives compare?
The 11 alternatives split into three groups. Personal agents from big companies (ChatGPT Work, OpenAI Dots, Gemini Spark, Claude, Grok Bot, Perplexity Teammates, Manus) work for one person. Lindy is one AI employee per person plus a shared one in Slack. OpenClaw and Hermes are free but self-hosted. Vestra is a team of agents run by a chief.
Here is each one in a line, by what it is, where you use it and where pricing starts. Prices are from each company's own pages on 14 November 2026, in US dollars unless noted.
Team of agents that work together
- Vestra: a chief that runs teammates on different engines (Vestra's own, Claude Code, Codex). Web, Mac app, Slack, Telegram and email. From $19 a month for up to 3 teammates; billed in rupees with GST in India.
Personal agents from big AI companies
- ChatGPT Work: one task at a time, in ChatGPT. Desktop, web and mobile. Limited on Free and Go ($8); fuller on Plus ($20).
- OpenAI Dots: an always-on agent in ChatGPT. Comes with ChatGPT Pro (from $100) or a Business Premium seat.
- Gemini Spark: a 24/7 background agent in Gmail, Docs and Sheets. For Google AI Pro and Ultra subscribers, including in India.
- Claude: Claude Code and scheduled tasks on paid plans. Pro $20 ($17 a month billed yearly), Max $100, Ultra $200.
- Grok Bot: Bots you create and brief one by one. Included with X Premium ($30), Super ($100) or Ultra ($200).
- Perplexity Teammates: agents that own ongoing work, in research preview. Perplexity Max, $200 a month.
- Manus: a general-purpose agent. Free, Pro and Team plans; prices are not listed on its plans page.
One AI employee
- Lindy: a private assistant per person, plus a shared Lindy in Slack, with its own US or Canadian phone number. Team from $29.99 a month per active user.
Free and self-hosted
- OpenClaw: free, MIT-licensed assistant you run on your own machine or server. Around 30 chat channels. You pay your model provider.
- Hermes Agent: free, MIT-licensed agent from Nous Research that runs where you put it. 20+ platforms. Works with many model providers.
And how they compare on the six things from the checklist, in short:
- Offered outside the US: the open-source ones work anywhere you can run them. Claude lists India and Canada as supported. Google brought Gemini Spark to India. OpenAI launched ChatGPT Go in India first. For the others, check the company's own pages for your country; several do not publish a country list on the pages we read.
- Own phone line: Lindy (US or Canadian number) and Vestra (your chief's own number, calls from Pro). OpenClaw can place calls through your own Twilio, Telnyx or Plivo account.
- Own email address: Vestra (your chief's own address) and Grok Bot (one address at mail.grokbot.com). Most others work in your own inbox once connected.
- Caps and receipts per task: Vestra caps any single task and the whole day and puts every task on a receipt. Most personal agents meter usage in weekly or five-hour allowances instead.
- Agents working together: Vestra (teammates on different engines, run by a chief). Others coordinate copies of themselves: Claude Code agent teams (experimental), Codex subagents, Grok Bot's primary Bot, Perplexity Teammates in a shared Project, Hermes Bot Mode, OpenClaw's team preset.
Which Meta Muse alternatives can you use outside the US?
You can use all 11 alternatives in this guide outside the US, though features and prices vary by country. Each section below says what the alternative is, where it works, what it costs, who it suits and the catch. Every competitor fact is from the company's own pages, checked on 14 November 2026, unless marked as a report.
1. Vestra: a team of agents run by a chief
What it is. Vestra is one office for your people and your AI agents. You name a chief, an agent that runs your other agents, speaks for you and answers only to you. You hire teammates by describing a job in a sentence, like "I need someone to get us paid on time". Your chief asks what a good manager would ask, then writes a teammate card: the job in one line, goals you can check, the apps it will use, its first routines and what it must ask you about first. You read it, change anything and say yes.
Why it is different from Muse. Muse is one agent for one person. Vestra is a team. Each teammate holds a whole job and runs on the engine that suits it: Vestra's own, Claude Code or Codex today. Your chief hands out the work, makes sure results are checked before they reach you, and only brings you real decisions. Correct one teammate and the lesson goes into the team's memory, so the others follow it too.
What it can do today. From Vestra's handbook, live today for one person and their team of agents:
- a chief and a team of teammates, with routines and a morning brief;
- a cloud computer with a real browser, which you can watch live and take over;
- calls and email from your chief's own number and address, with phone calls from the Pro plan up;
- more than 2,000 apps connected in a click, like Gmail, Slack, HubSpot, Notion, Shopify and Stripe;
- a memory that learns: a wiki of what your team knows, with a source for every line;
- any engine per teammate: Vestra's own, Claude Code or Codex, picked by you or your chief;
- limits, approvals and receipts: daily and per-task caps, your yes on money, every credit itemised;
- your chief in the app on the web and the Mac desktop app, in Slack, in Telegram and by email.
What is still being built. The office you can walk around, a chief for every person on a team, chiefs working together, spaces, meetings with agents taking notes, access control and WhatsApp and iMessage are being built now. A phone and inbox for every agent (not only the chief), a vault with a card per agent, shared apps, a shared brain for the whole company and working with clients chief to chief come after that. Vestra's site describes the product as in private beta in 2026.
Where you can use it. You sign up directly. Prices are in US dollars, and in India plans are billed in rupees with GST included. Every new office starts with 1,000 free credits.
Price. Lite is $19 a month ($17 billed yearly) for up to 3 teammates and 2,000 credits. Pro is $97 ($87 yearly) for 20 teammates, 10,000 credits and phone calls. Max is $197 ($177 yearly) for 50 teammates and 20,000 credits. Ultra is $497 ($447 yearly) for 150 teammates and 50,000 credits. One credit is one cent, the same for model time, the team's computer, searches and calls. Unused credits roll over.
Best for. Small businesses, founders and solo operators who want several jobs done, not one assistant, and who want to see what everything cost.
The catch. Several team features are still being built, including the walkable office and WhatsApp. Today's chief runs one person's team. If all you want is a personal helper for life admin on your phone, a personal agent may be simpler.
See the feature-by-feature Vestra vs Meta Muse comparison.
2. ChatGPT Work: one task at a time, inside ChatGPT
What it is. ChatGPT Work is a mode in ChatGPT for handing over a task with a clear outcome, like a brief, a deck or an analysis. You start each task, follow it in the chat and approve important actions. You can ask it to split independent parts across subagents that run in parallel and report back into the same chat. On Business and Enterprise, teams can share scheduled Team Tasks that run under a team service account.
Where you can use it. The ChatGPT desktop app on macOS, Windows and Linux, the web and the mobile app. OpenAI's own page says it introduced ChatGPT Go in India in August 2025, and that Go prices are localised in some markets.
Price. OpenAI's pricing page says Free and Go include limited ChatGPT Work on the desktop app. Plus adds expanded Work on desktop, web and mobile. Pro adds longer Work sessions and a dot. From our 14 November check of OpenAI's pricing docs: Go is $8 a month, Plus $20, Pro $100, $200 or $500, and Business $20 per user a month billed yearly ($25 monthly, 2 or more users). Work shares usage limits with Codex.
Best for. People who already live in ChatGPT and want one finished piece of work at a time.
The catch. It is still one person handing over one task. Subagents report back into your chat, not to each other on a standing basis, and everything runs on OpenAI's models. The docs we read do not describe a phone number or email address of its own. For the full contrast, see Vestra vs ChatGPT Work.
3. OpenAI Dots: an always-on agent with ChatGPT Pro
What it is. A dot is an always-on agent inside ChatGPT. You start with one primary dot that takes on ongoing work and brings results back for your review. OpenAI says specialist dots, which take on dedicated jobs inside a company, are starting with focused enterprise pilots.
Where you can use it. ChatGPT on the web, mobile and desktop, plus Microsoft Teams and Slack. Texting a dot over iMessage or RCS is a limited beta for Pro users in the US, Canada and Mexico.
Price. It comes with ChatGPT Pro at $100, $200 or $500 a month, or with a ChatGPT Business Premium seat.
Best for. Heavy ChatGPT users who want one agent that keeps working in the background.
The catch. One primary dot per person, and OpenAI says a dot can't have its own email address at launch. It is a personal agent with a high entry price. See Vestra vs OpenAI Dot.
4. Gemini Spark: Google's background agent, now in India
What it is. Gemini Spark is Google's background agent. Google describes it as a 24/7 personal AI agent that handles time-consuming digital tasks for you and keeps working even when your laptop is closed or your phone is locked. It runs on Google's cloud and is powered by Gemini 3.6 Flash.
Where you can use it. Google's India blog, dated 29 July 2026, says Spark was "officially expanding to Google AI Pro subscribers in India over the next few weeks" and is available to Google AI Ultra and Pro subscribers. It connects to Gmail, Docs and Sheets with no setup.
Approvals. Google says Spark is designed to ask you first before high-stakes actions like spending money or sending emails, and that you choose which apps it connects to.
Price. Included with Google AI Pro and Ultra subscriptions. Google's post does not give rupee prices.
Best for. People and teams whose work already lives in Google Workspace, and who are in India today.
The catch. It is a personal agent, centred on Google's own apps and Google's own model. If your business runs on tools outside Google, or you want different agents for different jobs, it is a narrower fit.
5. Claude: Claude Code and scheduled tasks on paid plans
What it is. Claude is Anthropic's assistant. Paid plans include Claude Code, Anthropic's coding agent, plus research, Claude in Chrome and the ability to hand off and schedule tasks. Team plans list Claude Cowork alongside Claude Code.
Where you can use it. Anthropic's supported-countries page lists India and Canada. Claude Code runs in the terminal, editors, the desktop app, the web and the mobile apps.
Price. Claude's pricing page lists Free, Pro at $20 a month ($17 billed yearly), Max at $100 and Ultra at $200. Team seats are $25 a month ($20 yearly) for Standard and $125 ($100 yearly) for Premium, for 2 to 150 seats. Prices are in US dollars and the page says they don't include tax.
Best for. Developers and writers who want a strong model and a capable coding agent.
The catch. Claude Code is built for one developer at a time. Its agent teams, where a lead session assigns work to other sessions, are experimental and off by default, and every worker is a Claude session. In Vestra, Claude Code can be one teammate's engine next to Codex and Vestra's own. See Vestra vs Claude Code.
6. Grok Bot: Bots you create and brief one by one
What it is. xAI calls Grok Bot "your proactive AI teammate", an agent with access to a real computer in the cloud. You create each Bot yourself, give it a job, sources and approval limits, and your primary Bot coordinates the others.
Where you can use it. Desktop apps for macOS, Windows and Linux, iPhone and Android, Slack, and by tagging @bot on X. xAI's overview page does not state country availability, so check before you subscribe.
Price. Included with an X subscription: Premium at $30 a month, Super at $100 or Ultra at $200. Bot usage comes from a weekly pool shared with the Grok apps.
Email and calls. You can claim one email address at mail.grokbot.com for your Bots. You can talk to a Bot on a voice call, but xAI's docs don't describe Bots calling other people.
Best for. People who already pay for X and want agents with a computer of their own.
The catch. Each Bot is briefed one by one, and they all run on xAI's models for one account. See Vestra vs Grok Bot.
7. Perplexity Teammates: agents that own ongoing work
What it is. Perplexity's help centre says "Teammates are AI agents that own ongoing work". You give one a responsibility, like triaging bug reports, and it plans the tasks and carries them through to a result for your review. They are built on Perplexity Computer and can watch connected tools like Slack, GitHub and Linear.
Where you can use it. On the web at perplexity.ai. The help article does not list regional limits.
Price. The research preview needs an individual Perplexity Max subscription, $200 a month or $2,000 a year. Enterprise Pro and Enterprise Max are not included by default.
Best for. Research-heavy people who already pay for Perplexity Max.
The catch. It is a research preview, and Perplexity says behaviour may change. It is priced for heavy individual users, and you set up each teammate yourself. See Vestra vs Perplexity Teammates.
8. Lindy: one AI employee, with a phone line
What it is. Lindy describes itself as one AI employee. Each person gets a private Lindy for their inbox, meetings and calendar, and the team shares one Lindy it mentions in Slack. Lindy says most teams give it several roles at once.
Phone and email. Lindy can have its own US or Canadian phone number that makes and takes calls. Email replies are drafted in your voice and sent from your own inbox.
Price. Lindy's pricing page, checked on 14 November, says every active user in a workspace is billed, from $29.99 a month for 3,000 credits each, with top-ups at $10 per 1,000 credits. Monthly credits don't roll over. (On 13 November the same page showed $49.99 for 5,000 credits, so check it before you buy.)
Best for. Teams that want one assistant in Slack and email, especially in the US or Canada where its phone numbers work.
The catch. It is one agent wearing many hats, and it is billed per person. The phone numbers are US or Canadian, which matters if your customers are in India or the UK.
9. Manus: a general-purpose agent with team credits
What it is. Manus is a general-purpose AI agent that takes on tasks on a computer of its own.
Price. Manus's plans page lists Free (limited monthly credits), Pro (a larger monthly credit allocation and priority support) and Team (a shared team credit pool, collaboration and admin controls). It does not list prices on that page, and the pricing page did not show them when we checked.
Best for. People who want a capable single agent and a shared credit pool for a team.
The catch. A shared credit pool is not the same as agents that work together. Each task is still one person handing work to one agent. Because prices were not shown on Manus's own pages when we checked, we can't compare its cost.
10. OpenClaw: free, open source and self-hosted
What it is. OpenClaw is a free, open-source assistant under the MIT licence, stewarded by the OpenClaw Foundation. You install it and run its Gateway yourself. One command adds a team preset: a chief-of-staff coordinator plus a researcher, a writer and a reviewer.
Where you can use it. Anywhere you can run it. It reaches around 30 chat channels through plugins, including WhatsApp, Telegram, Slack, Discord, Signal, iMessage and Microsoft Teams.
Price. No subscription. You pay your model provider and for the machine or server that runs the Gateway.
Best for. Technical people who want full control and are happy to run and secure a server.
The catch. You are the IT department. You host it, update it, hold the keys and decide who can trust the shared Gateway. For a business, that is real work. Read OpenClaw alternatives for businesses and Vestra vs OpenClaw.
11. Hermes Agent: Nous Research's self-improving agent
What it is. Nous Research calls Hermes Agent "the self-improving AI agent", one that gets more capable the longer it runs. It keeps a memory that grows across sessions, searches past sessions and builds a model of the user.
Where you can use it. Its docs say it lives wherever you put it, on a $5 VPS, a GPU cluster or serverless infrastructure, and reaches 20+ platforms from one gateway, including Telegram, Discord, Slack, WhatsApp, Signal, email and SMS.
Teamwork. It has a Bot Mode, described as a durable team of specialist Bots that work together in group chats, and it can spawn isolated subagents for parallel work.
Price. Free under the MIT licence. It works with Nous Portal, OpenRouter, OpenAI or any endpoint, so you pay whichever model provider you choose.
Best for. Builders who want an agent that runs on their own server and talks on WhatsApp or Telegram.
The catch. Like OpenClaw, you run and secure it yourself, and its Bots are Hermes Bots on your one server. There is no hosted office, no one receipt across engines and nobody to call when it breaks.
Which Meta Muse alternative is best for a small business in India?
For a small business in India, the best Meta Muse alternative depends on how many jobs you need done. For one assistant, Gemini Spark (if you run on Google Workspace) or ChatGPT Work are offered in India today. For several jobs at once, like sales follow-up, customer replies and invoices, Vestra gives you a team run by a chief, billed in rupees with GST.
Most small businesses in India do not have one job that needs help. They have six. Think of a 12-person export business in Jaipur, a dental clinic in Bengaluru or a D2C brand in Pune. The owner is doing sales, chasing payments, answering customers, posting on Instagram, checking suppliers and keeping the books, often in a mix of WhatsApp, Gmail, Tally or Zoho, and a spreadsheet.
A personal agent helps that owner with one thing at a time. It is a better assistant. It is not a team. Here is how the options fit, by what the business needs:
- "I need help with my own inbox and calendar." Gemini Spark if you are on Google Workspace and a Google AI Pro or Ultra subscriber. ChatGPT Work if you already use ChatGPT.
- "I need someone to write content and research." ChatGPT Work, Claude or Perplexity all do this well for one person.
- "I need several jobs handled, and I don't want to manage five agents." Vestra. Your chief hires a teammate for each job, like a Sales teammate to follow up leads, a Customer care teammate to answer customers and a Bookkeeping teammate to send invoices and chase late payments. Vestra comes with ready-made skills for Sales, Bookkeeping, Content, Customer care and Research.
- "I have a developer and want full control." OpenClaw or Hermes, on a server you run.
Things that matter more in India than the US:
- Rupee billing with GST. Vestra plans are billed in rupees with GST included in India. Most US agents charge in dollars, though OpenAI says some ChatGPT Go prices are localised. A rupee invoice with GST is simpler for your accountant.
- WhatsApp. Indian businesses run on WhatsApp. We cover this in the next section, because the honest answer is mixed.
- Your own tools. Check that the agent connects to the tools you actually use. Vestra connects 2,000+ apps, and where there is no connector, a teammate can use the website on its own cloud computer, the way a person would.
- Approvals. For a business, the agent must never spend money, delete things or message a new customer without a yes. In Vestra those three always wait for a person, and no setting turns that off.
- Cost control. You cap spending on any single task and on the whole day, up to $100 a day. Anything unusually big, over 50 credits by default, waits for your okay with the cost shown.
A simple way to decide: list the three jobs that eat most of your week. If one person could do all three, a personal agent may be enough. If they are really three different jobs, you want a team. For more on this, read The best AI agents for business.
Which alternative is best if your business runs on WhatsApp?
If your business runs on WhatsApp, the self-hosted agents OpenClaw and Hermes connect to it today, and Muse itself works in WhatsApp where it is offered. Vestra's chief works in Slack, Telegram, email and its own app today; WhatsApp and iMessage are being built now. Check each tool's WhatsApp support before you choose.
Here is the honest picture, from each company's own pages:
- Meta Muse: you can message Muse in WhatsApp, but only where Muse is offered: the US, Canada and Singapore.
- OpenClaw: WhatsApp is one of its chat channels, through a plugin you set up on your own Gateway.
- Hermes Agent: WhatsApp is one of the 20+ platforms its gateway reaches, on the server you run.
- Vestra: your chief works in the Vestra app on the web and Mac, in Slack, in Telegram and by email today. WhatsApp and iMessage are in Vestra's "building now" list. Telegram is live, if your team can use it in the meantime.
There are two kinds of "WhatsApp support", and it helps to keep them apart:
- Talking to your agent on WhatsApp. You message your assistant from your phone. This is about your convenience.
- Your agent talking to customers on WhatsApp. The agent replies to customers or suppliers. This is about your business, and it is where approvals matter most, because a wrong message to a customer is hard to take back.
For the second kind, look for an agent that asks before contacting anyone new, keeps a record of every message and lets you take over. In Vestra, contacting someone new for the first time always waits for a person's yes, whatever channel the work runs through.
Which alternative is best for developers?
For developers, Claude (with Claude Code) and Codex (inside ChatGPT) are the strongest coding agents, and both can be used outside the US. If you want them to work together, one writing code and one reviewing it, Vestra lets each be the engine under a different teammate, run by one chief. If you want to host everything yourself, OpenClaw and Hermes are free.
Developers have the most options outside the US, because coding agents mostly need a terminal, a repository and an API key, not local shops or payment partners.
- Claude Code: included in every paid Claude plan, from Pro at $20 a month. Supported in India and Canada, per Anthropic's list.
- Codex: comes with ChatGPT plans, from limited access on Free and Go to more on Plus, Pro and Business.
- OpenClaw and Hermes: free and open source, and both can use Claude Code, Codex or other models underneath.
The problem developers hit is not finding a good coding agent. It is getting two of them to work together. Claude Code coordinates Claude Code sessions. Codex coordinates Codex subagents. Neither can review the other's work on its own, so you end up as the glue.
In Vestra, each can be the engine under a different teammate. Vestra's handbook gives the example: one teammate builds on Claude Code and another reviews the work on Codex, so a different engine reads the code with fresh eyes, and the chief keeps everyone posted. Both work under the same office rules, and the receipt shows what each one spent. We wrote a whole guide on this: How to use Claude Code and Codex together. See also Vestra vs Claude Code.
Which Meta Muse alternatives are free or cheapest?
The free options are OpenClaw and Hermes Agent, which cost nothing to use but need your own server and model provider, plus the free tiers of ChatGPT, Claude and Manus, which have tight limits. The cheapest paid options are ChatGPT Go at $8 a month and Claude Pro or ChatGPT Plus at about $20. Vestra starts at $19 a month for a team of up to 3 teammates.
Here is the ladder from cheapest up, from each company's own pages:
- $0 (self-hosted): OpenClaw and Hermes Agent. Free software. You pay for the model and the server.
- $0 (free tiers): ChatGPT Free (limited Work on desktop), Claude Free (no Claude Code), Manus Free (limited monthly credits).
- About $8 a month: ChatGPT Go, with limited Work on desktop. OpenAI says Go prices are localised in some markets.
- About $19 to $20 a month: Vestra Lite at $19 ($17 billed yearly), Claude Pro at $20 ($17 billed yearly) and ChatGPT Plus at $20. Muse Power is $20 in the US.
- About $30 a month: X Premium, which includes Grok Bot, and Lindy Team at $29.99 per active user.
- $100 a month and up: ChatGPT Pro (with a dot), Claude Max, X Super, Muse Maximum (in the US).
- $200 a month: Perplexity Max (for Teammates), Claude Ultra, X Ultra.
Price alone hides two things.
First, what you get for it. $19 for Vestra Lite gets you a chief and up to 3 teammates, 2,000 credits and an 8 GB computer for your team. $20 for a personal agent gets you one agent. Neither is better in every case, but they are not the same thing.
Second, what a task actually costs. Many personal agents meter use in weekly allowances or five-hour windows, so you find out you are out of usage when the agent stops. Vestra uses credits where one credit is one cent, and every piece of work gets a receipt that shows model time, actions and computer time, and which teammate spent it. Vestra's handbook gives an example of a whole Tuesday of work for a small team costing 393 credits, or $3.93.
How do you pay for an AI agent in rupees?
To pay for an AI agent in rupees, pick one that bills in rupees or offers local prices. Vestra plans are billed in rupees with GST included in India. OpenAI says some ChatGPT Go prices are localised. Most other agents in this guide list US dollar prices, so an Indian card is charged in dollars. Muse has no confirmed Indian price yet.
What we found, company by company:
- Vestra: prices are in US dollars, and in India plans are billed in rupees with GST included.
- ChatGPT: OpenAI's own page says it introduced ChatGPT Go in India in August 2025 and that Go prices are localised in some markets. The pricing page we read showed US prices.
- Google AI Pro and Ultra: Google brought Gemini Spark to Indian subscribers; its post does not give rupee prices.
- Claude: prices are listed in US dollars and do not include tax.
- Grok Bot, Perplexity, Lindy: prices are listed in US dollars on the pages we read.
- Meta Muse: not offered in India. Reported rupee prices (₹1,799 for Power and ₹8,999 for Maximum) are unconfirmed by Meta.
For a business, three practical points:
- GST invoices. If you claim input tax credit, you want an invoice that shows GST properly. Ask before you buy.
- Card limits and foreign charges. Dollar subscriptions on Indian cards can hit international limits or carry foreign transaction fees.
- Usage-based billing. Know what happens when you run out. In Vestra, unused credits roll over to next month and you can top up any time without changing plans. Every new office starts with 1,000 credits on the house.
How do you run several AI agents together instead of one at a time?
To run several AI agents together, put one agent in charge, give each of the others a whole job, keep one shared memory and one set of rules, and count work as done only when it has been checked. Personal agents can't do this with each other. A team product like Vestra is built for it: your chief runs teammates on different engines.
This is where the Muse question turns into a bigger one. Say you give up waiting for Muse and use three alternatives instead: ChatGPT Work for writing, Gemini Spark for your inbox, Claude Code for your website. You now have three assistants who never talk to each other. You are their manager. You brief each one, carry work between them, chase every result and pay three bills.
The way out is structure. These are the same things that make a human team work:
- One lead. One agent that knows your goals, hands out the work and is the only one that talks to you. In Vestra, that is your chief.
- Whole jobs. Each teammate owns a job with goals you can check, not a stream of one-off prompts. "Get us paid on time" is a job. "Send this invoice" is a task.
- One memory. Every teammate reads the same facts, with a source for each one. You brief once.
- One set of rules. Spending over a limit, deleting something and contacting someone new always wait for a person. Everything else is tuned teammate by teammate.
- Checked work. Done means checked: by a test, a second agent, the source or a person.
- One receipt. Every piece of work shows what it cost and who spent it.
For a full guide, read How to manage multiple AI agents without becoming their manager.
In Vestra, those six things come with the office. And because a teammate's engine sits underneath its job, you can switch engines without losing anything: its job, goals, skills, apps, routines and memory belong to the teammate, not the engine. If Muse's engine becomes a real option one day, you would not start again. Muse appears in Vestra's engine picker in the handbook, but the engines live today are Vestra's own, Claude Code and Codex.
What does a week look like with one personal agent, and with a team?
With one personal agent, a busy week means you brief it on every task, copy its results into other tools and check everything yourself. With a team led by a chief, you get a morning brief with the few decisions that need you, and the rest is done, checked and on a receipt. Here is the same week, both ways, for a small business outside the US.
Picture Riya. She runs a small business in Bengaluru that sells appointment software to clinics. This is a made-up example to show the difference, not a customer story. She has three jobs she wants off her plate: following up with leads, answering customer questions and getting invoices paid.
The personal-agent way
Riya can't get Muse in India, so she uses a mix of personal agents she can get.
- Monday. She asks her chat agent to draft follow-ups to eight leads. She pastes in the lead list, explains her prices again and copies the drafts into Gmail one by one.
- Tuesday. A customer asks a question her inbox agent can't answer, because it doesn't know what the sales agent promised. She digs through Monday's chat to find out.
- Wednesday. She asks a different agent to list unpaid invoices. It doesn't know which clients always pay late, because that was in last month's conversation with another tool. She explains it again.
- Thursday. One agent reaches its weekly usage limit in the middle of a task. She waits, or upgrades.
- Friday. She checks every draft, every reply and every reminder herself, because no agent checked another's work. She has three bills, three logins and three memories, none of which agree.
Each agent did good work. Riya was the glue.
The team way
Now the same week in Vestra. Riya named her chief Asha. She hired three teammates in a sentence each: a Sales teammate, a Customer care teammate and a Bookkeeping teammate.
- Monday, 8 am. Asha's morning brief arrives in Slack: what got done, the one thing that needs Riya and what Asha noticed on its own. One decision: the Sales teammate wants to email a clinic it has never contacted. Riya replies "1" to approve. First contact always waits for her.
- Tuesday. A customer asks about a discount the Sales teammate offered. The Customer care teammate already knows, because the offer is in the team's memory, with the email it came from as its source.
- Wednesday. The Bookkeeping teammate sends invoices on its routine and drafts a friendly reminder to the two clients the memory says usually pay late. Nobody explained it again.
- Thursday. The Sales teammate wants to buy a contact list. It costs more than the per-task cap Riya set, so it stops and asks, with the cost shown.
- Friday. Riya reads one receipt for the week: which teammate spent what, on which job, in credits where one credit is one cent.
Riya made three decisions all week. That is the difference between a better assistant and a team. Note what is live today: one person's team, the chief, teammates, memory, limits and receipts, and Telegram. Chiefs for every colleague, the walkable office and WhatsApp are being built.
How do you switch from Meta Muse, or stop waiting for it?
To switch from Meta Muse, or stop waiting for it, write down what you wanted it to do, pick an alternative that is offered in your country, connect the tools those jobs need and start with one familiar job so you can check the results. If you used Muse before, copy what matters from its Memory.md file first.
A simple plan for this week:
- List the jobs, not the features. Write down the three things you wanted Muse to do. "Answer customer emails within two hours" is better than "email".
- Decide: one assistant or a team? If it is one job for you, pick a personal agent from the list. If it is several jobs, or you want agents to check each other's work, pick a team.
- Check it is offered where you are. Confirm sign-up, payment and the features you need, like calls or a local app, work in your country.
- Bring your context. If you used Muse in a supported country, Meta lets you view and edit what it remembers in Memory.md. Copy what matters. In Vestra, the memory also reads the history of the tools you connect on day one, so it already knows your customers.
- Set the rules first. Decide what must always wait for you, and set spending caps before the agent touches money.
- Start with one familiar job. Something you know well enough to judge the result. In Vestra, tell your chief what you need in a sentence and read the teammate card before you say yes.
- Loosen the rules as trust grows. In Vestra, new teammates start by asking first, then suggest, then act and tell you, but only when you move the setting.
If Muse launches in your country later, test it next to what you are using, on the same job. You will know quickly whether it does more for you.
What is live in Vestra today, and what is still being built?
Live in Vestra today, for one person and their team of agents: a chief and teammates, a cloud computer, calls and email from your chief, 2,000+ connected apps, a memory that learns, an engine per teammate (Vestra's own, Claude Code or Codex), limits, approvals and receipts, and your chief in the app, Slack, Telegram and email.
Vestra's handbook keeps a plain list, called Where We Stand, of what works now and what comes next. Here it is, so you know exactly what you are buying.
Live today, for one person
- A chief and a team of teammates: hire, shape and run teammates, with routines and a morning brief.
- A cloud computer: watch your team work in a real browser, and take over any time.
- Calls and email from your chief: its own number and address, with phone calls from Pro up.
- Connected business tools: email, calendar, CRM, books and chat, connected in a click.
- A memory that learns: a wiki of what your team knows, with a source for every line.
- Any engine, per teammate: Vestra's own, Claude Code or Codex, picked by you or your chief.
- Limits, approvals and receipts: daily and per-task caps, your yes on money, every credit itemised.
- Your chief in the app, Slack, Telegram and email, on the web and the desktop app for Mac.
Building now
- A chief for every person on a team, named by them.
- Chiefs working together: ask, answer, hold or say no, and hand work across.
- The office you can walk around: every desk, every space and what is happening, live.
- Meetings in the office, with agents keeping the notes.
- Spaces: one goal, shared context and decisions for the right person.
- Access control: private, company or network, for every teammate, space and app.
- Every hand-off checked and written down, with one record for both sides.
- WhatsApp and iMessage.
Then, one at a time
- A phone and an inbox for every agent, not only the chief.
- The vault: a card and logins for every agent, sealed, with limits each person sets.
- Shared apps: software built in a sentence, used by people and agents together.
- The shared brain: one brain for the whole team, learning from every correction.
- The front door: clients and partners working with you, chief to chief.
Vestra says some live features are still rolling out to every account, and its site describes the product as being in private beta in 2026. Gemini, Grok and Muse appear in the engine picker in the handbook's pictures, but the engines listed as live are Vestra's own, Claude Code and Codex.
Frequently asked questions
Is Meta Muse available in India?
No, not as of 14 November 2026. Meta's own pages list the US and Canada (for adults) and Singapore. Indian press reports say Indian users are on a waitlist and that Meta has listed rupee prices, but Meta has not confirmed a launch date or prices.
When is Meta Muse coming to India?
Meta has not given a date. A Business Upturn report says Meta's India head told Moneycontrol that bringing Muse to India would be a priority next year, while the company weighs how to manage the rollout.
How much will Meta Muse cost in India?
Unconfirmed. Analytics India Magazine, citing The Hindu, reported on 10 November 2026 that Meta has listed Power at ₹1,799 a month and Maximum at ₹8,999, with app store billing costing slightly more. In the US, Meta's Help Center lists Power at $20 a month and Maximum at $100.
Which countries is Meta Muse available in?
The US and Canada, for people 18 and over, according to Meta's Muse for Small Business page, and Singapore since 27 October 2026, according to Meta's newsroom. Some features, like voice and calls to businesses, started in the US only.
Is Meta Muse available in the UK, Europe or Australia?
Not as of 14 November 2026. Meta's pages name only the US, Canada and Singapore, and say Meta plans to expand to more markets without giving timing.
What is the best Meta Muse alternative?
It depends on the job. For one personal agent, ChatGPT Work and Gemini Spark are closest and are offered in many more countries. For coding, Claude Code. For a team of agents that work together under one chief, one memory and one set of limits, Vestra, from $19 a month.
What is the best free alternative to Meta Muse?
OpenClaw and Hermes Agent are free and open source, but you run them on your own server and pay your model provider. ChatGPT, Claude and Manus have free tiers with tight limits. Vestra gives every new office 1,000 free credits to try real work before picking a plan.
Can I use Meta Muse on WhatsApp in India?
Muse works in WhatsApp where it is offered, which does not include India yet. A press report says Muse's web and WhatsApp interfaces can be reached from India with features restricted, but Muse is not officially available there. Vestra's chief works in Telegram, Slack, email and its app today; WhatsApp is being built.
Is Vestra available in India?
Yes. You sign up directly, and in India Vestra plans are billed in rupees with GST included. Prices are set in US dollars, from $19 a month for Lite. Every new office starts with 1,000 free credits.
How is Vestra different from Meta Muse?
Muse is one personal agent for one person. Vestra gives you a chief that runs a team of teammates, each holding a whole job on its own engine: Vestra's own, Claude Code or Codex. They share one memory and one set of rules, and every piece of work has a receipt. See Vestra vs Meta Muse.
Can I use Claude Code and Codex in Vestra?
Yes. You pick the engine for each teammate: Vestra's own, Claude Code or Codex. One teammate can build on Claude Code while another reviews on Codex, both run by your chief under the same approvals, limits and receipts.
Does Vestra ask before it spends money?
Yes. Spending over your limit, deleting something and contacting someone new always wait for a person's yes. You cap any single task and the whole day, up to $100 a day, and anything unusually big waits for your okay with the cost shown.
Can an AI agent have its own phone number outside the US?
Some can. Vestra's chief has its own number and email address, with phone calls from the Pro plan up. Lindy offers US or Canadian numbers. OpenClaw can place calls through your own Twilio, Telnyx or Plivo account. Muse's calls to businesses are, per a report, for paid US users only.
Should I wait for Meta Muse or use an alternative now?
If you need help this quarter, use an alternative now. Meta has not given a date for most countries, and features often arrive later than the app. If Muse launches where you are, test it on the same job and compare.
Glossary
AI agent. Software that does tasks for you, not just answers questions. You give it a goal and it takes the steps, in apps and websites.
Personal agent. An AI agent built for one person. Its memory, history and bill belong to that person, and it works alone. Muse, ChatGPT Work, Gemini Spark, Claude Code, Codex, Grok Bot, OpenAI Dots and Hermes are personal agents.
AI team. Several agents with separate jobs, run by one lead agent, sharing one memory and one set of rules.
AI office. A shared place where people and their agents work as one team. Vestra is an AI office.
Chief. In Vestra, the agent that runs your other agents, speaks for you and answers only to you. You name it.
Teammate. In Vestra, an agent hired for a whole job with goals, shaped by you, on the engine you pick.
Engine. What a teammate runs on: the part that does the thinking and the working. In Vestra, Vestra's own, Claude Code or Codex today.
Muse Spark. Meta's model that powers Muse.
Muse Secure VM. The dedicated cloud computer Muse works in, which holds the agent and your data.
Muse tokens. The weekly allowance Muse's paid plans are measured in.
Credit. In Vestra, the unit you pay for work in. One credit is one cent, the same for model time, the computer, searches and calls.
Receipt. A record of what a piece of work cost, who did it and who said yes.
Self-hosted. Software you run on your own machine or server, like OpenClaw and Hermes Agent, rather than a service someone runs for you.
Approval. A yes from a person before an agent acts. In Vestra, spending over a limit, deleting something and contacting someone new always need one.
Waitlist. A list of people waiting for a product to open in their country. Indian users are reportedly on Muse's.
GST. India's Goods and Services Tax. Vestra plans in India are billed in rupees with GST included.
Sources
- Meta: Introducing Muse, the world's first personal AI agent built for everyone
- Meta: Muse for Small Business
- Muse for Small Business page
- Meta: Muse is now available in Singapore
- Meta: Give your Muse a voice
- Meta Help Center: About Muse subscriptions
- Meta Help Center: Muse memory and data
- Analytics India Magazine: Meta to introduce Muse AI agent in India, report
- Business Upturn: Meta lets paid Muse users call US businesses, with India launch still pending
- ChatGPT pricing
- OpenAI: Introducing ChatGPT Go (Hindi)
- ChatGPT docs: Get started with Work
- ChatGPT docs: Pricing
- OpenAI Help: Getting started with your dot
- Google: Introducing Gemini Spark in India
- Claude pricing
- Anthropic: Supported countries and regions
- Claude Code docs: Agent teams
- xAI: Grok Bot overview
- xAI: X Subscription pricing
- Perplexity Help Center: Get started with Perplexity teammates
- Perplexity Help Center: Perplexity Max
- Lindy pricing
- Lindy docs: Phone line
- Manus docs: Plans
- OpenClaw
- OpenClaw docs: Channels
- Hermes Agent docs, Nous Research
