biondizzle 7409204d71 Use nightly's deepseek_v4.py + attention as base, add only NVFP4 mapper ...

The upstream deepseek_v4.py has imports that don't exist in the nightly
Docker image (norm_gate_linear, breakable_cudagraph, etc.). Use the
nightly's own files as the base and add only the minimal NVFP4 changes:
- Add _make_deepseek_v4_nvfp4_weights_mapper() for checkpoint key mapping
- Select NVFP4 mapper when quant_config is modelopt_fp4
- cos_sin_cache float32 fix in attention
- Remove utils.py patch (not needed)

2026-05-18 22:33:51 +00:00

..

patches

Use nightly's deepseek_v4.py + attention as base, add only NVFP4 mapper

2026-05-18 22:33:51 +00:00

cutedsl_quant_method.py

Fix: add abstract create_weights to CuTeDSLNvfp4LinearMethod

2026-05-18 20:40:48 +00:00

nvfp4_cutedsl.py

Fix torch.compile: use custom autograd Function instead of @torch.compiler.disable

2026-05-18 21:38:28 +00:00