biondizzle 3f2f4e1882 Fix cudaErrorStreamCaptureUnsupported: no dynamic GPU-tensor slicing ...

Dynamic slicing with GPU scalars (e.g. buf[:gpu_scalar]) is a CUDA
operation not permitted during stream capture. Use full pre-allocated
buffers instead of dynamic slices. The GEMM only reads rows indicated
by expert_offsets, ignoring the zero padding.

Also pass x_sf[:num_slots] (Python int slicing, cudagraph-safe) to
scale assembly so it only processes real token scale data.

2026-05-17 17:24:26 +00:00

..

patches

Fix shape mismatch: shared padded buffers, revert max_num_tokens cap

2026-05-17 15:47:10 +00:00

nvfp4_cutedsl.py

Fix cudaErrorStreamCaptureUnsupported: no dynamic GPU-tensor slicing

2026-05-17 17:24:26 +00:00