The new model, Muse Glimmer, is smaller than leading AI models from rivals and is instead designed to run agentic tasks on a Mac or PC with a single graphics card, aiming to tap demand for AI systems that run directly on people's devices.
"We've got even bigger models that are coming soon," Zuckerberg said in a video post accompanying his 14-page essay titled "The Future is for Everyone" in which he championed spreading AI rather than leaving it in the hands of a few.
"The notion AI is so dangerous that the only safe path is an extreme concentration of power seems inherently problematic."
Zuckerberg's statement marks the latest show of support for open-weight AI, which is gaining traction as businesses grow wary of ballooning AI bills and worry about recent cybersecurity incidents involving models from Anthropic, OpenAI and Meta.
Open-weight models are typically cheaper than leading models from the likes of OpenAI and Anthropic. They also come with publicly accessible core components for easy customization, unlike closed models that companies keep fully under their control.
Those controls are in the spotlight after Hugging Face, the AI coding collaboration hacked by a rogue OpenAI model, was forced to use a Chinese open-weight model to defend against the attack because closed-source models have curbs on cybersecurity use.
Chinese startups are leading the race for open-weight models, with Moonshot's Kimi K3, Alibaba's Qwen3.8-Max and DeepSeek's V4-Flash rivaling the performance of top U.S. systems. By contrast, the leading models from U.S. developers OpenAI, Anthropic and Alphabet's Google are closed-weight.
Shares of Meta, which have fallen about 10% so far this year, were up nearly 3% in premarket trading on Monday.
POLICY RETHINK NEEDED TO PROPEL OPEN-WEIGHT
Meta also plans to release the weights of Muse Spark 1.2, its most advanced model, built by a costly superintelligence team it formed last year to claw its way back into the AI race.
The company had long supported open models and was among the only U.S. tech firms to release such models before poor reception for its Llama 4 last year forced a change in strategy.
In Monday's wide-ranging AI essay, Zuckerberg also unveiled a new $1 billion fund that aims to ease fears about the impact of its data-center build-out by offering support to communities.
Growing local opposition to data centers has become a key election issue in the U.S. and one of the biggest hurdles for Big Tech's push to build the infrastructure needed to power AI.
"One significant disadvantage that the U.S. has compared to countries like China is that it is more difficult to build infrastructure here," Zuckerberg, whose company is set to spend as much as $145 billion this year on AI infrastructure, said.
Zuckerberg also said the U.S. needed to rethink policies if domestic firms were to lead in open-weight models, including on data use and distillation - a method that uses the outputs of a powerful AI system to train a smaller model that can handle some of the same tasks with far less computing power.
He said that Meta would implement a governance structure to give its independent directors the power to approve the safety criteria for releasing models.
"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.
"U.S. policy must reduce this additional friction if we want American open source models to lead over time," Zuckerberg said, adding that restricting access to foreign open-source models was not an effective solution.
President Donald Trump's administration told AI developers earlier this month that it will not put open-weight AI models through voluntary safety tests, according to two sources familiar with the discussions.
