trellis-2-mrp-mlx/scripts
m3ultra 7860148eb2
Some checks are pending
CodeQL Advanced / Analyze (${{ matrix.language }}) (none, c-cpp) (push) Waiting to run
CodeQL Advanced / Analyze (${{ matrix.language }}) (none, python) (push) Waiting to run
spconv Metal kernel (opt-in) + fix vertex baker scheduler fallthrough
- sparse_conv_metal: fused gather-GEMM kernel, parity 8e-4 vs stock;
  honest verdict: stock chunking already bounds memory at decoder scale
  (1.97GB synthetic peak) and beats the scalar kernel on speed — kept as
  TRELLIS2_METAL_SPCONV=1 opt-in + upstream reference, NOT default.
- _attempt_schedule: preferred=vertex now runs the vertex baker (was
  silently falling through to the 5h+ pure-python kdtree grind).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-20 03:37:20 +10:00
..
download_weights.py fix: cache only pinned runtime weights 2026-07-17 09:24:41 +03:00
generate_asset.py spconv Metal kernel (opt-in) + fix vertex baker scheduler fallthrough 2026-07-20 03:37:20 +10:00
probe_macos.py feat: add native Apple Silicon asset pipeline 2026-07-17 09:21:34 +03:00
setup_macos.sh fix: retry Metal build with ABI fallback 2026-07-17 09:26:00 +03:00
validate_outputs.py feat: add native Apple Silicon asset pipeline 2026-07-17 09:21:34 +03:00