Meta Launches Muse Wonder With a Memory for the Plans You Hoped Would Blow Over
Meta's new model family pairs cross-conversation memory with persistence on difficult tasks. Wonder is available through the Meta Model API and Muse Code; its arrival in Muse comes later.
Automation · · 3 views
Remember the business idea you enthusiastically explained before deciding that opening another tab was enough entrepreneurship for one evening? A fresh chat used to offer a certain emotional distance from that version of yourself.
Meta introduced Muse Wonder and Muse Wonder Full on October 12. The company describes Active Memory as carrying relevant details across conversations, alongside Active Focus, which is designed to keep working through difficult tasks, recover from errors and seek clarification when necessary. Wonder is available through the Meta Model API and Muse Code, with Full available to Muse Code subscribers. Wider Muse integration is still to come. The pitch is continuity: fewer explanations repeated, more work carried forward. I can see the appeal of retaining your requirements, although retaining your enthusiasm may prove a separate engineering challenge.
Meta says the models are designed to check in before important actions. That leaves a useful opening for you to explain that the ambitious project is now a casual interest, and that the casual interest has become something you would prefer not to discuss. An assistant that remembers the original brief can at least make the change of scope unusually well documented. The app may remain unbuilt, but its cancellation could finally receive your full attention.
Based on: Introducing Muse Wonder and Muse Wonder Full: Frontier Models for Agents, Meta, October 12, 2026.