Meta Says Muse Spaces Turn Planner and Tracker Prompts Into Pages That Update in the Background
On November 6, 2026 Meta introduced spaces in its Muse app, custom pages built from a text description that update in the background from connected apps, and began rolling out specialized Muse and Meta AI models trained using its Muse Wonder frontier series.
Meta is turning its Muse personal agent from a chat window into something users can rearrange, adding “spaces”: custom pages that Muse builds from a plain-language description and then keeps up to date on its own. The company introduced the feature on November 6, 2026, and said spaces are available starting that day everywhere Muse is offered. Alongside it, Meta began rolling out specialized versions of the models behind Muse and Meta AI, trained using its Muse Wonder frontier series.
The two announcements fit together. Spaces gives Muse a place to keep the results of ongoing work, and the new models are meant to remember context and keep working on a goal.
From a sentence to a page
Meta describes a space as “a custom experience you create inside Muse by describing your idea.” A user might ask for a holiday gift tracker, a daily word game or a planning board for a DIY project. Muse builds it, and the user can keep refining it from there.
Meta’s own examples lean on its other services and on connected apps:
- a shopping hub that pulls in gift ideas, wishlists from Instagram saves, and the user’s Marketplace listings and buyer messages;
- a school dashboard that finds events relevant to a user’s children in school emails and adds them to a calendar;
- an order tracker for a small business, showing new orders, customer messages and what needs to ship that day;
- a trip planner with flights, hotel, reservations and a day-by-day itinerary.
Pages that update in the background
What separates a space from a saved chat is that it is meant to stay current. Meta says that once users connect the apps they use, “your spaces keep themselves up to date in the background,” so new concert dates, price changes or savings progress appear without a new prompt. Users can check items off, add notes and edit directly inside a space.
Spaces also change the layout of the app itself. Users can pin a space to the sidebar, move it into the main tabs, hide it, or schedule when it appears. Meta’s example is a weekly planner kept up front during the workweek, then swapped for weekend plans on Saturday. The company says it is also exploring ways to let people share spaces and work on them together, which it has not yet shipped.
Models trained with Muse Wonder
Meta introduced Muse Wonder and Muse Wonder Full on October 12, 2026, calling them its most capable models for personal and professional agentic tasks. At the time it said Muse Wonder was available through the Meta Model API and Muse Code, and would later power Muse, starting with subscribers.
The November 6 update takes a different route. Instead of switching Muse to Muse Wonder directly, Meta says it trained specialized versions of its existing models using Muse Wonder and is rolling them out to people who use Muse and Meta AI. These models bring over two features first introduced with Muse Wonder:
- Active Memory, which Meta says helps the assistant tell temporary information from permanent information, reason across time, and spot when an earlier detail becomes relevant again.
- Active Focus, which Meta says helps the assistant keep working toward a goal when it hits an unexpected problem by trying harder, recovering, or finding another route.
Meta’s own scorecard
All of the figures Meta published come from its internal evaluations. On what it calls the Personal Superintelligence Index, the current Muse model scores 47%, up from 38% for the previous one. On an Active Memory evaluation measuring tasks that correctly use critical context, the score rose from 76% to 90%. Separately, Meta says the improved models in Muse show a 2.4x reduction in forgetting critical information. Meta also reports a 67% improvement in completing everyday tasks where unexpected issues arise.
Meta has not published the methodology behind these benchmarks, and none of the numbers has been independently checked. The company also says the models have learned to judge when to push on a goal independently and when to collaborate live with the user, and to handle mid-task changes of direction “without getting confused.”
App support and refresh intervals remain unspecified
Meta’s announcement does not list which outside apps can feed a space, how often background refreshes run, or whether spaces count against any usage limits. It also does not say which specialized model version a given user is getting, or how the rollout differs between Muse and the Meta AI chatbot. For now, spaces are a single-user feature, with sharing described only as something Meta is exploring.
- #Meta
- #Muse
- #Muse Wonder
- #AI Agents
- #Personal AI
Author
Raj M
Contributor
AI Systems Architect is a seasoned technology leader with over 15 years of experience in the IT industry working with Fortune 500 companies. With a solid foundation in multi-agent systems, open-source LLM infrastructure, and enterprise deployment, he excels at building scalable production-grade AI platforms.