Home · Technology · Oct 12 archive
Meta Launches Muse WonderClaude
Confirmed
In Short: Meta has launched the Muse Wonder AI model, a follow-up to its earlier Muse Spark models, as part of its ongoing efforts to revitalize its artificial intelligence division.
Meta has launched the Muse Wonder AI model, a follow-up to its earlier Muse Spark models, as part of its ongoing efforts to revitalize its artificial intelligence division.
The company's CEO, Mark Zuckerberg, announced the new model during a recent post on X, stating that the release had been delayed to ensure safety.
According to ProPakistani, both versions of the Muse Wonder model support a 1 million token context window and can process text, images, video, and audio.
The new model is available through an API and through Muse Code, a tool for developing software with AI.
Meta priced Muse Wonder at $4 per million input tokens and $20 per million output tokens, positioning it competitively in the market.
The company has also released a smaller open-weight model called Muse Glimmer and plans to release larger open-weight models in the future.
Zuckerberg overhauled Meta's AI research last year by forming the Meta Superintelligence Labs and hiring Alexandr Wang, the former CEO of Scale AI, to lead the effort.
The new lab's first model, Muse Spark, was announced in April, and the latest release marks a significant step forward in Meta's AI ambitions.
Meta emphasizes the model's 'active memory' and 'active focus' features, which allow it to retain and reuse user information over extended periods and continue working through challenges.
The company's efforts to ensure the safe scaling of its models are evident in its recent blog post detailing a framework designed to manage the models' capabilities.
Despite these advancements, Meta's AI glasses continue to face privacy concerns and worries about nonconsensual filming and surveillance, similar to the issues that plagued Google Glass.
What this adds
Meta's Muse Wonder model is part of a broader strategy to regain competitive footing in the AI race, following a significant investment in its AI division last year.
The pricing of Muse Wonder aligns with other high-end AI models like GPT-6 Sol and Claude Opus 5.5, indicating a competitive market for advanced AI capabilities.
Meta's approach to scaling and safety reflects broader industry concerns, as seen in OpenAI's response to a cybersecurity breach, which included pausing reinforcement learning training and scaling efforts.
Background
Meta has launched its MuseOpenAI's ChatGPT on the iPad, expanding its availability beyond the iPhone and Mac.
What's confirmed
- Meta has launched the Muse Wonder AI model, a follow-up to its earlier Muse Spark models, as part of its ongoing efforts to revitalize its artificial intelligence division.
- The company's CEO, Mark Zuckerberg, announced the new model during a recent post on X, stating that the release had been delayed to ensure safety.
- According to ProPakistani, both versions of the Muse Wonder model support a 1 million token context window and can process text, images, video, and audio.
- The new model is available through an API and through Muse Code, a tool for developing software with AI.
- Meta priced Muse Wonder at $4 per million input tokens and $20 per million output tokens, positioning it competitively in the market.
- The company has also released a smaller open-weight model called Muse Glimmer and plans to release larger open-weight models in the future.
- Zuckerberg overhauled Meta's AI research last year by forming the Meta Superintelligence Labs and hiring Alexandr Wang, the former CEO of Scale AI, to lead the effort.
- The new lab's first model, Muse Spark, was announced in April, and the latest release marks a significant step forward in Meta's AI ambitions.
- Meta emphasizes the model's 'active memory' and 'active focus' features, which allow it to retain and reuse user information over extended periods and continue working through challenges.
- The company's efforts to ensure the safe scaling of its models are evident in its recent blog post detailing a framework designed to manage the models' capabilities.
- Despite these advancements, Meta's AI glasses continue to face privacy concerns and worries about nonconsensual filming and surveillance, similar to the issues that plagued Google Glass.
What's still developing
- Meta launched the Muse series after an expensive shake-up within its AI division last year that brought the launch… 9:53 AM ET Dow Jones Futures: Micron and Nvidia stock led the AI trade lower Monday.
- Meta's AI glasses are grappling with privacy worries and concerns about nonconsensual filming and surveillance, the same pushback that derailed category pioneer Google Glass. (© Jon Krause) Get instant access to exclusive stock lists, expert market analysis and powerful tools with 2 months of IBD Digital for only $20!
- TRENDING: Meta Leads In AI Glasses, But Still Faces 'Creep' Hurdle Meta Platforms (META) stock climbed Monday after Chief Executive Mark Zuckerberg introduced an AI model that he says puts the Facebook parent company at the frontier.
- It certainly suggests that Meta is once again competitive in the increasingly high-stakes AI race.
- Meta’s latest AI models, Muse Wonder and Muse Wonder Full, seem to show that the bet is finally paying off by putting the company back at the frontier of AI development.
- Meta says the best version of the model will power features across its platforms, including its Muse personal agent.
- Its Llama series of models were the first frontier open-weight offerings from a big lab.
- Muse Wonder Full boasts frontier capabilities, Meta says, meaning that it is among the most capable AI models in the world.
- OpenAI is changing how it trains, monitors and contains frontier AI models after its July cybersecurity evaluation breach reached Hugging Face's production infrastructure.
- "Our largest planned frontier RL run remains on hold while we conduct smaller-scale training and evaluations to assess model behavior, validate our safeguards, and establish more evidence of alignment before proceeding."
- OpenAI said that immediately after the Hugging Face incident it also paused frontier-model inference in research clusters for workloads capable of executing code or using tools that could reach the internet.
- The company said it temporarily slowed the pace of scaling, paused reinforcement learning training, and is continuing to hold its largest planned frontier RL run while it puts stronger security and alignment controls in place.