5 Commits
Author SHA1 Message Date
poxularandCursor f223858ac4 Add GPU Fast pipeline for sub-3s H100 inference.
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-06-02 01:16:07 +02:00
poxularandCursor 4965f27ba4 Add full GPU pipeline with ComfyUI nodes and phase profiling.
Ports scene extract, face ONNX preprocess, and net reference resize to Torch/CUDA with minimal CPU sync, while keeping the CPU parity path unchanged.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-06-02 01:01:46 +02:00
poxularandCursor d101418eb9 Add separate CUDA ONNX path for experimental GPU inference.
Keeps CPU parity nodes and providers unchanged; CUDA modules and ComfyUI
nodes run alongside with CPU fallback when CUDA EP is unavailable.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-06-02 00:22:25 +02:00
poxularandCursor a251a0ed9f Add live strength slider: auto_refresh queues Preview while Build stays cached.
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-06-01 17:00:32 +02:00
poxularandCursor 4d2d3d4236 Initial release: MAGICMATCH ComfyUI color-match nodes with bundled ONNX model.
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-06-01 16:51:00 +02:00