Popular repositories Loading
-
hotskip
hotskip PublicRun MoE LLMs bigger than RAM + VRAM on a laptop: hot experts in VRAM, expert cache in RAM, the rest from SSD. Patch + CLI for llama.cpp.
Python
-
soberstrap
soberstrap PublicBloxstrap-like mods, graphics filters and themes for Sober (Roblox on Linux)
Python
-
mousedpi
mousedpi PublicTiny DPI / report rate / RGB settings app for gaming mice on Linux (libratbag front-end)
Python
-
hotskip-image
hotskip-image PublicHunyuanImage-3.0 (80B MoE, needs 3-8x80 GB) on an 8 GB laptop GPU: per-layer streaming of experts, ~78 s per 1024x1024 image
Python
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.