biondizzle e46b615873 Fix dense router BF16 dispatch for CUDA graph capture ...

- Run GEMM in BF16 (not FP32) during graph capture — Blackwell tensor cores
  handle BF16 natively; FP32 GEMM triggers cudaErrorStreamCaptureUnsupported
- Pre-transpose W_gate to [E, H] at load time — avoids .T view during capture
- Convert only logits output to FP32 for sqrt(softplus) numerical stability
- This fixes the graph capture failure at layer 0 Graph B

2026-06-04 05:50:13 +00:00

..

attention

Wire prefill FMHA into production.py and single_shot

2026-06-03 03:49:57 +00:00

cache

Cleanup Step 2: Archive Lineage P code, fix broken imports

2026-06-02 19:27:07 +00:00

compressor

Cleanup Step 2: Archive Lineage P code, fix broken imports

2026-06-02 19:27:07 +00:00

cuda

Add c10/cuda/CUDAStream.h include for getCurrentCUDAStream

2026-06-04 04:13:40 +00:00

gemm

Blackwell swizzle CUDA kernel for CUDA graph capture

2026-06-04 03:03:02 +00:00

indexer

Cleanup Step 2: Archive Lineage P code, fix broken imports

2026-06-02 19:27:07 +00:00

router

Fix dense router BF16 dispatch for CUDA graph capture

2026-06-04 05:50:13 +00:00

__init__.py

Restructure: cutedsl/ -> dsv4/ with proper layering

2026-05-21 17:30:44 +00:00