Muse Wonder
- ID
- 34637
- Status
- summarized
- Published
- 13 Oct 2026, 12:48 AM
- Fetched
- 13 Oct 2026, 3:46 AM
- Provider
- Hacker News
- Category
- dev-community
- Original URL
- https://research.meta.ai/blog/muse-wonder
- Source URL
- https://hnrss.org/best
Summary
- Score
- 4.0
- Created
- 13 Oct 2026, 3:46 AM
- Tags
- Audience
- developersai_ml_learnersai_agent_userssaas_startup_founders
What happened
Meta AI Research introduced the Muse Wonder series — Muse Wonder and Muse Wonder Full — describing them as its most capable models for personal and professional agentic tasks, built around two named features, Active Memory and Active Focus. Both are said to be available today through the Meta Model API and Muse Code, with Muse Wonder Full gated to Muse Code subscribers, and the post says the models were trained at Meta's gigawatt-scale Prometheus data center. The excerpt contains no benchmark scores, pricing, context limits, or regional availability, and the HN thread drew 275 points and 124 comments.
Why it matters
If you already build on the Meta Model API or Muse Code, there is a new model tier to check — but the only concrete change stated is that Muse Wonder Full sits behind a Muse Code subscription while the base Muse Wonder is reachable via the API, and no pricing or eval numbers are published here to compare against your current model spend. Practically, that means you cannot make a cost or capability decision from this post alone; you'd have to hit the API or the evaluation report to get anything usable. Nothing in the text touches Malaysia-specific availability, data residency, or pricing, so treat local relevance as unknown rather than implied.
Discussion angle
A vendor claiming state-of-the-art agentic performance with zero numbers in the launch post — what would you actually test on Active Memory and Active Focus before migrating a long-running agent off your current model, and does a 275-point HN thread change your mind either way?