Jack Paul 
Meta Launches Two New Muse Models
Meta has introduced two artificial intelligence models, Muse Wonder and Muse Wonder Full. The company calls them its most capable models so far and says they move toward what it describes as personal superintelligence. Meta announced the models on Monday.
The pair adds two features Meta calls Active Memory and Active Focus. Active Memory is meant to carry relevant information from one conversation to the next. Active Focus helps the agent stay on difficult tasks without losing the thread. Those are practical features, though the company has not shared full technical details yet.
Benchmark Scores Draw Praise and Doubt
Independent evaluation platform Artificial Analysis scored Muse Wonder Full at 56 on its Intelligence Index. The standard Muse Wonder scored 52. Both scores came at maximum reasoning effort. That places Wonder Full above Claude Fable 5.1 and GPT-6 Astra, which each scored 53. It still sits below Claude Opus 5.5, which scored 58.
The results quickly drew pushback. Bindu Reddy, CEO of Abacus.AI, runs her own model evaluations. She wrote on X that Muse Wonder is nowhere near Fable or Astra level. Reddy said detailed benchmark results would follow. Her criticism does not disprove Meta’s numbers, but it does show how contested model rankings remain. Different tests, prompts, and settings can produce different winners. For now, buyers should treat single index scores as a starting point, not a final answer.
Pricing and Access
Meta’s new models are its most expensive to date. The standard Muse Wonder charges $4 per million input tokens. That matches Claude Opus 5.5 and undercuts GPT-6 Astra at $10 and Claude Fable 5.1 at $50. For teams watching costs, that middle position may matter as much as benchmark bragging rights.
Muse Wonder Full is not open to everyone. It is available only to people who subscribe to Muse Code’s $50 a month power usage tier or its $200 a month maximum usage tier. That limits who can test the top-end model directly. It also means many public comparisons will likely focus on the cheaper standard version first.
Meta is making a strong claim about its position in the AI race. The early scores support part of that claim, but not all of it. Rival labs still have models at or above Meta’s best result. The next few weeks should bring more independent tests, user reports, and cost comparisons. Until then, Muse Wonder looks competitive, expensive at the top end, and not yet clearly ahead.