diff --git a/scripts/build_prequant_checkpoint.py b/scripts/build_prequant_checkpoint.py index 8312e02130..366822de94 100644 --- a/scripts/build_prequant_checkpoint.py +++ b/scripts/build_prequant_checkpoint.py @@ -86,7 +86,9 @@ def main(argv = None) -> int: quantize_( transformer, _make_quant_config(scheme), - filter_fn = make_filter_fn(args.min_features, exclude_name_tokens = exclude_tokens_for_scheme(scheme)), + filter_fn = make_filter_fn( + args.min_features, exclude_name_tokens = exclude_tokens_for_scheme(scheme) + ), ) # Move the state dict to CPU for a portable, GPU-free artifact. diff --git a/studio/backend/core/inference/diffusion_transformer_quant.py b/studio/backend/core/inference/diffusion_transformer_quant.py index e9bd1c6ff1..821f33da1b 100644 --- a/studio/backend/core/inference/diffusion_transformer_quant.py +++ b/studio/backend/core/inference/diffusion_transformer_quant.py @@ -72,6 +72,7 @@ def exclude_tokens_for_scheme(scheme: str) -> tuple[str, ...]: M=1 projections as int8 and crashes at the first denoise step on Flux / Qwen.""" return _INT8_EXCLUDE_NAME_TOKENS if scheme == TQ_INT8 else () + # Per-architecture preference order for ``auto`` -- best (fastest, in-bar) first, with # the lower-precision schemes listed as fallbacks for that arch tier. On Blackwell, fp8 # leads: measured on a B200, plain fp8 dynamic is both faster AND more accurate than the