AI models · Six models, from 2 GB to 200 GB
Each Mac lends Feya about two thirds of its memory, so you can keep using it. Models past the dashed line are split across several Macs.
Qwen 2.5 1.5B Short, simple questionsFits on any Mac
SD-Turbo Pictures from a text descriptionFits on any Mac
Llama 3.1 8B Emails, summaries, ideasFits on a 16 GB Mac
Gemma 4 31B Harder questions that need reasoningA big Mac, or two together
Llama 3.3 70B Long, detailed answersSeveral Macs together
DeepSeek V4 Flash The most capable model on FeyaMany Macs together
You don’t have to choose. Your Mac reports how much memory it can spare (part of its RAM, capped by what the GPU can handle), and the coordinator gives it a model that fits, or makes it a helper on a bigger one. A small starter model downloads first, and a better one follows in the background.
The coordinator picks the largest model the online Macs can hold between them. One Mac leads, the others help, and the model’s layers are split by how much memory each one has. The lead runs the model; the helpers compute their layers over llama.cpp’s RPC protocol. It works best when the Macs are close to each other on the network.
What runs on your Mac