Projects
Mergen — Distributed LLM Inference on Reclaimed Phones
ActiveDistributed C++20/GGML LLM inference runtime running 1.1B to 27B models (Qwen3.5, Qwopus 9B, Qwen3.8-27B, LFM3, Xing4.0 MoE) across reclaimed Linux and Android phones.
Distributed C++20/GGML LLM inference runtime running 1.1B to 27B models (Qwen3.5, Qwopus 9B, Qwen3.8-27B, LFM3, Xing4.0 MoE) across reclaimed Linux and Android phones.