Meta CEO Mark Zuckerberg used Monday’s model launch to do two things at once: release a new AI model and make a public case for U.S. policy reform.
Introducing Muse Glimmer, an open-weight 30B-parameter model optimized for local, always-on agent workflows.
Muse Glimmer delivers strong performance on key agentic use cases and benchmarks compared with leading models in its size category, and is designed to run entirely on… pic.twitter.com/mI4z91GPnE
— AI at Meta (@AIatMeta) August 10, 2026
The new model is called Muse Glimmer. It is designed for agentic tasks and is small enough to run on a Mac or PC with a single graphics card. That puts it in a different category from the large frontier models coming out of OpenAI, Anthropic, and Google.
Meta stock was up 0.37% on Monday as the announcement dropped.
The launch comes after Meta formed a new superintelligence team last year, a costly move aimed at getting the company back into the front of the AI race. Muse Glimmer appears to be part of that wider push.
Open-weight models give users access to the core components of the model, making them easier and cheaper to customize than closed-source alternatives. Businesses have been warming to them as AI costs climb and security concerns around closed models grow.
Hugging Face pointed to this directly last month. The coding collaboration platform was hacked by a rogue OpenAI model and said it used a Chinese open-weight model to defend against the attack, citing restrictions on closed-source models for cybersecurity work.
Zuckerberg acknowledged that open-weight models are where the competitive pressure is coming from. Chinese labs are currently ahead. Moonshot’s Kimi K3, Alibaba’s Qwen3.8-Max, and DeepSeek’s V4-Flash are all delivering performance that competes with top U.S. systems.
“Foreign labs currently hold several advantages here since American labs have to comply with many additional restrictions on training data,” Zuckerberg said.
His argument is straightforward. U.S. companies face extra compliance burdens that Chinese labs do not. If that gap does not close, American open-source models will fall further behind.
“US policy must reduce this additional friction if we want American open source models to lead over time,” he said. He added that blocking access to foreign open-source models would not fix the problem.
Zuckerberg also backed AI model distillation, a process where a powerful model is used to train a smaller one. It is a technique that makes building capable lightweight models more efficient.
The Trump administration told developers earlier this month it will not require open-weight AI models to go through voluntary safety tests, according to sources familiar with the discussions. That is a policy direction that aligns with what Zuckerberg is asking for.
On governance, Meta said it plans to give its independent directors the authority to approve safety criteria before new models are released.
Meta said it plans to launch more open-weight models in the near future.
The post Meta Stock: Zuckerberg Launches Muse Glimmer and Calls for Open-Source AI Policy Reform appeared first on CoinCentral.
