cmoyates d13ba1639d feat: add optimization toggle flags + benchmark matrix script ... use_sdpa, stage_gc flags propagated through backbone/model/pipeline. bench_optimizations.py runs exhaustive/ablation/key sweep of all 6 toggles (slim, stage_gc, sdpa, bf16, fused_decode, gpu_preprocess) with latency + peak memory reporting. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>		2026-03-09 17:11:28 -02:30
..
bench_mlx.py	feat(phase6): optimization, benchmarking, and tiled inference	2026-03-01 06:24:10 -03:30
bench_optimizations.py	feat: add optimization toggle flags + benchmark matrix script	2026-03-09 17:11:28 -02:30
compare_reference.py	feat(phase6): optimization, benchmarking, and tiled inference	2026-03-01 06:24:10 -03:30
convert_weights.py	feat(phase3): PyTorch→MLX weight converter (#1)	2026-03-01 05:16:56 -03:30
dump_pytorch_reference.py	fix: match original CorridorKey decoder concat order [c4,c3,c2,c1]	2026-03-01 22:58:08 -03:30
infer_pytorch.py	chore: fix lint, format, type errors	2026-03-08 21:29:06 -02:30
infer.py	fix: resize inputs to model img_size, use infer_and_save in CLI	2026-03-01 06:11:57 -03:30
smoke_2048.py	feat: add 2048 smoke test for native-resolution validation	2026-03-01 07:04:28 -03:30
smoke_engine.py	feat: add CorridorKeyMLXEngine integration surface	2026-03-01 06:43:07 -03:30