Muse Wonder is available through the Meta Model API and Muse Code, while Muse Wonder Full is available to Muse Code subscribers

Paid AI assistants now offer advanced features like proactive task completion and improved error handling.

Meta has launched its new AI model, Muse Wonder, which it claims is its most advanced tool for powering digital assistants. Access to Muse Wonder requires a subscription to the Muse personal agent app, with monthly plans priced at $20 or $100. This release signifies Meta's strategy to generate revenue from AI services beyond its traditional online advertising business. According to Meta AI chief Alexandr Wang, the model, internally codenamed Watermelon, is designed to enhance the capabilities of personal and professional agents, enabling them to handle tasks like booking reservations and managing availabilities more effectively by exploring alternative solutions when encountering obstacles. The paid version of the model includes enhanced safeguards against misuse, a feature that is not available in the free version. Muse Wonder will also be accessible through Meta's Model API for developers, at a price point Wang stated will be lower than comparable models from competitors like OpenAI and Anthropic.

The introduction of Muse Wonder and its tiered subscription model highlights a broader industry trend where advanced AI capabilities are increasingly being monetized through direct user fees, shifting away from purely ad-supported or free models. This move by Meta suggests a growing market appetite for sophisticated AI agents that can perform complex real-world tasks, and a willingness among some users to pay for enhanced functionality and security. The company's focus on improving agent resilience and offering stronger safeguards indicates a response to user feedback and potential concerns about AI reliability and safety, positioning them to capture a segment of the market seeking dependable AI tools.

Meta's strategy with Muse Wonder positions it to compete directly in the emerging market for AI-powered personal and professional agents, challenging established players. By offering a more advanced model with enhanced features, they aim to attract users who require higher performance for tasks that digital assistants have historically struggled with. The emphasis on safeguards also addresses growing concerns around AI misuse, which could be a key differentiator. This development could spur further innovation in agentic AI, with competitors likely to respond by improving their own models' capabilities and safety features, potentially leading to a more robust and diverse AI assistant ecosystem.