.png){kind=logo}
Meta introduced a new open-weight model as CEO Mark Zuckerberg pushes for a wider shift toward systems that can be downloaded, customized and operated directly on personal computers. The move comes as the social media company tries to strengthen its position in a fiercely competitive technology market while US and Chinese developers race to advance their own models.
The newly released Muse Glimmer is considerably smaller than the most advanced models offered by companies such as OpenAI and Anthropic. It can run on a Mac or PC equipped with a single graphics card, potentially making it more accessible to developers and businesses that want to process tasks on their own hardware.
Unlike closed systems, open-weight models provide access to key components that developers can customize for particular applications. They can also be less expensive to operate, an increasingly important consideration as businesses face growing costs associated with advanced computing services.
For Meta, the strategy could create another route to wider adoption of its technology. Smaller systems that operate locally can appeal to developers and companies that want greater control over their data, computing infrastructure and software deployments.
Zuckerberg has also indicated that Meta intends to release additional open-weight models in the future, signalling that the latest launch is part of a longer-term strategy rather than a one-off product release.
Alongside the launch, Zuckerberg called for changes to US policy, arguing that American companies face additional restrictions that could make it harder for them to compete with overseas developers in the open-weight market.
Chinese companies have recently gained ground in this segment, with models from Moonshot, Alibaba and DeepSeek attracting attention for their performance. Meanwhile, several leading US developers, including OpenAI, Anthropic and Google, continue to operate predominantly closed systems.
“Foreign labs currently hold several advantages here since American labs have to comply with many additional restrictions on training data,” Zuckerberg said, referring to open-source models.
“US policy must reduce this additional friction if we want American open source models to lead over time,” Zuckerberg added.
Also Read: Meta Ordered to Pay $567 Million and Introduce New Teen Safety Measures
The shift towards smaller, locally operated systems also reflects changing priorities among businesses. Companies are increasingly looking for technology that can be customized, deployed on their own infrastructure and operated at a manageable cost.
Meta also plans to introduce a governance framework under which its independent directors would have authority to approve safety criteria before new models are released. Zuckerberg additionally backed model distillation, a process in which a smaller system learns from a more capable one.
Meta shares, which have fallen approximately 10% so far this year, were up about 1% in premarket trading on Monday (August 10, 2026) following the announcement.
