This app allows you to run large language models locally by pooling the RAM of multiple iOS and macOS devices. Connect devices via Wi-Fi or wired connections to unlock models that exceed single-device capacity. It supports both tensor and pipeline parallelism for efficient inference.
0.3.1
May 19, 2026
- Added Gemma4 models - Bugfixes and improvements
Infer Ring is a free application that enables local LLM inference by pooling the RAM of multiple iOS and macOS devices. It allows you to run models that would not fit on a single machine by combining their memory.
You run the app on each of your iOS and macOS devices to pool their RAM. Devices can connect over Wi-Fi or a wired USB connection to utilize their combined processing power for LLM inference.
Infer Ring supports both tensor parallelism and pipeline parallelism. This allows for efficient distribution of large language models across multiple devices.
Yes, Infer Ring is available for free. There are no in-app purchases or subscription fees associated with its use.
Infer Ring is compatible with macOS, iPhone, and iPad devices. You can combine the memory of these Apple devices to run LLMs.
The latest version, 0.3.1, was released on May 19, 2026. While specific update frequency isn't detailed, the app is actively being developed.
No, Infer Ring does not contain advertisements. As a free application with no in-app purchases, it provides an ad-free experience.
Infer Ring has an age rating of 4+. This indicates it is suitable for a wide range of users, including younger individuals.
