Science & Tech · AI · 4 hrs ago
Meta Introduces Paid Muse Wonder AI Models
Meta has introduced two paid AI models, Muse Wonder and Muse Wonder Full, as it tries to strengthen its position in the AI race.
The models build on Muse Spark, which Meta launched earlier this year.
Meta says Muse Wonder can help with tasks such as reviewing lease agreements, handling insurance claims and building apps.
It adds Active Memory, which can carry information between conversations, and Active Focus, which can try other approaches when it hits a problem.
Meta says the model is designed to protect private information and resist cyberattacks, though storing information across conversations raises privacy concerns.
Muse Wonder is currently available to developers through Meta’s Model API plans, which cost $20 or $100 a month.
Meta says it will soon use the model in Muse experiences with pricing based on usage.
The launch comes as Meta competes with AI companies including Anthropic, OpenAI and Google, amid calls for stronger safety measures and disagreement over whether companies can police themselves.
Meta debuted its Muse Wonder and Muse Wonder Full premium AI models on Monday.
The models build on Muse Spark and are described by Meta as its most capable models to date.
Muse Wonder is initially available to coders through paid Meta Model API plans.
Meta says the models can handle tasks including insurance claims, lease agreements and building apps.
The launch comes amid debate over AI safety, regulation and whether companies can police themselves.
- Who
- Meta introduced Muse Wonder and Muse Wonder Full.
- What
- It launched two premium AI models, with Muse Wonder initially offered through paid API plans.
- When
- Monday; the article was published on October 12, 2026.
- Where
- Not stated.
- Why
- Meta says the models extend its AI capabilities and can perform personal and professional tasks as it competes with other AI companies.
AI researchers and advocates for external oversight
Mark Zuckerberg and Meta
AI oversight
AI researchers and advocates for external oversight
Many researchers and experts advocate external evaluation and regulation of AI; some executives have called for slowing its development.
Mark Zuckerberg and Meta
Zuckerberg believes AI labs can and should police themselves.
Safeguards
AI researchers and advocates for external oversight
CNET’s Katelyn Chedraoui said safeguards will be essential for a new frontier model.
Mark Zuckerberg and Meta
Zuckerberg said AI can be engineered with normal guardrails and built responsibly.
Meta launched the Muse Spark AI model, which the new models build upon.
Meta debuted Muse Wonder and Muse Wonder Full.
- Models
- Muse Wonder and Muse Wonder Full
- Underlying model
- Muse Spark
- Power Plan
- $20 per month
- Maximum Plan
- $100 per month
- New features
- Active Memory and Active Focus
What people are saying
But this new frontier model comes amid serious safety concerns in the AI industry. Safeguards will be essential to any new launch.
I’m much closer to the belief that AI is a technology that can be engineered with normal guardrails and built responsibly than the anthropomorphic vision of AI, which is that it is just going to go do its own thing.
most capable models to date
Sources
Related news
Microsoft president calls for global AI safety rules
Meta unveils two AI models it says rival leading competitors
Anthropic AI submits false murder tip to Philadelphia police
Meta launches Muse Wonder models to compete with leading AI systems
Zhipu tests GLM-5.3-Highspeed with selected ZCode subscribers
CEATEC 2026 opens in Chiba with 821 exhibitors
Scholar urges Global South to move beyond supplying AI raw materials
Meta launches new AI models, with faster development planned
Robotics competitions in Hangzhou to feature embodied AI challenges
Samsung outlines AI memory strategy beyond faster HBM
Nadella calls for human-controlled emergency brakes on advanced AI
