Alibaba and Meta bring open AI models to laptops
The next AI battle may happen on your device, not in the cloud. Here's how Alibaba and Meta are making it possible.
AI is moving closer to the user. Alibaba and Meta are pushing open-weight AI models that can run on personal computers, signalling a shift towards AI that does not always need a cloud data centre.
For developers and advanced users, this means capable models can increasingly be downloaded and run locally, provided their hardware is powerful enough.
Why open-weight AI matters
Open-weight models make a trained model's parameters available for users to download and run themselves. This gives developers more control to test, customise and deploy AI without depending entirely on a company's cloud service.
It is worth noting that open-weight does not always mean fully open-source. The model weights may be available while the training data, code or development process remains private. Running AI locally can have several advantages.
Prompts and files can remain on the device, reducing the need to send sensitive information to a cloud service. Local models can also work without an internet connection and may reduce ongoing API costs for some users.
Alibaba and Meta push smaller models
Alibaba's Qwen family has become a major player in the open-model ecosystem. The company offers models in different sizes, including Qwen3 variants designed for reasoning, coding and multimodal tasks.
Alibaba's documentation lists Qwen3 models ranging from 8B to much larger versions, giving developers options based on available computing resources.
The smaller models are particularly relevant to local AI because they require considerably less memory and processing power than the largest systems. This makes them more practical for developers using powerful laptops or desktop computers.
Meta is taking a similar approach with its Muse family. The company recently released Muse Glimmer as an open-weight model designed to run on personal devices, including Macs and PCs with a single graphics card.
Meta is also doubling down on its commitment to open-weight AI. The company is known for its Llama models, and CEO Mark has argued that AI should not be controlled by a small group of companies or governments.
What local AI could mean for users?
The shift could make AI assistants more personal and useful. A locally running model could help users write, summarise documents, code or organise information without sending every request to a remote server.
However, local AI still has limits.
Larger models require significant memory and computing power, while smaller models may not match the capabilities of the most powerful cloud systems. Battery consumption, speed, storage and model safety also remain practical concerns.
Even so, Alibaba and Meta are helping push AI beyond the data centre. As smaller and more efficient models improve, laptops could become an increasingly important part of the AI ecosystem, giving users greater control over where and how their AI tools run