biondizzle e1fcfc4f01 Add CuTeDSL warmup + CUDA sync after JIT compilation ...

CuTeDSL cute.compile corrupts GPU memory. Add warmup forward +
torch.cuda.synchronize() + health check after finalize_weights,
matching the MoE runner pattern.

2026-05-19 01:11:44 +00:00

..

nvfp4

Add CuTeDSL warmup + CUDA sync after JIT compilation

2026-05-19 01:11:44 +00:00