MacPaw is tapping Liquid AI to offer on-device inference to developers building for its app store, TechCrunch reports. The partnership is aimed at letting app makers add AI features that run locally rather than relying entirely on remote model calls.
On-device inference can improve responsiveness and privacy because user data does not always need to leave the machine. It can also reduce recurring server costs for developers whose AI features fit within local hardware limits.
Liquid AI has focused on efficient model architectures, which makes it a natural fit for apps that need smaller local models. For MacPaw, the move gives its developer ecosystem a clearer AI tooling story.
The constraint is capability. Local models can be useful for focused tasks, but developers still need to match model size, latency, and quality to what a user’s device can realistically handle.