Pinned Loading
-
minimax-h3-mac
minimax-h3-mac PublicRun MiniMax H3 (video + native stereo audio) locally on Apple Silicon — two ComfyUI workflows, measured cost tables, and the dead ends so you don't repeat them
-
MLX-DLSS
MLX-DLSS PublicForked from iamwavecut/MLX-DLSS
NVIDIA DLSS 5 neural rendering and DLSS frame generation on Apple Silicon (MLX/Metal, Core ML) and PyTorch; weights extracted from your own DLSS libraries
Swift
-
krea2-turbo-mac
krea2-turbo-mac PublicRun Krea 2 Turbo fully locally on Apple Silicon — ComfyUI workflows (bf16/MPS) with in-graph model downloads, the locked 16-step recipe, and every Mac-specific trap documented (batch=1, LoRA-0.0 bl…
-
ltx-faceid-lipsync
ltx-faceid-lipsync PublicIdentity-locked talking video from one photo + your own voice — ComfyUI workflows for LTX-2.3 Face-ID with frozen-audio lipsync (Mac + CUDA)
-
ideogram4-mac
ideogram4-mac PublicRun Ideogram 4 locally on Apple Silicon — two core-node ComfyUI workflows on the official recipe, measured costs, and the JSON caption grammar
-
h3image
h3image PublicMiniMax H3 as an image editor + generator: ComfyUI workflows and a CLI for edit, masked inpaint, detail, outpaint, text-to-image and face autofix. Apple Silicon (MPS) and CUDA.
Python
If the problem persists, check the GitHub status page or contact support.




