gbyu-amd cb7214d8ea

[ROCm][MLA] enable fp8 MLA decode on ROCm (#28032) ...

Signed-off-by: guanbao <gyu@amd.com>
Signed-off-by: Guanbao Yu <gyu@amd.com>
Signed-off-by: gbyu-amd <Guanbao.Yu@amd.com>
Co-authored-by: guanbao <gyu@amd.com>

2025-11-25 10:15:02 +08:00

..

mla

[ROCm][MLA] enable fp8 MLA decode on ROCm (#28032)

2025-11-25 10:15:02 +08:00

__init__.py

…

cpu_attn.py

[perf][cpu] Accelerate paged attention GEMMs (QK, PV) on Arm CPUs with NEON (#29193)

2025-11-22 09:04:36 -08:00

flash_attn.py

[Attention] Refactor FA block_size limitations to hybrid models only (#29084)

2025-11-22 06:38:44 -08:00

flashinfer.py

[Attention] Refactor FA block_size limitations to hybrid models only (#29084)

2025-11-22 06:38:44 -08:00

flex_attention.py

[CI Failure] Fix backend selection for encoder-only models (#28534)

2025-11-13 10:11:27 -05:00

gdn_attn.py

[Perf] Refactor cudagraph_support to enable full CUDA graphs for spec decoding with FlashInfer (#28479)

2025-11-13 01:56:40 +09:00

linear_attn.py

…

mamba1_attn.py

Remove deprecated fields from CompilationConfig (#27593)

2025-11-12 16:10:28 +00:00

mamba2_attn.py

Remove deprecated fields from CompilationConfig (#27593)

2025-11-12 16:10:28 +00:00

mamba_attn.py

[Perf] Refactor cudagraph_support to enable full CUDA graphs for spec decoding with FlashInfer (#28479)

2025-11-13 01:56:40 +09:00

pallas.py

[V0 deprecation] Remove no longer used get_metadata_cls (#28370)

2025-11-10 14:32:09 +08:00

rocm_aiter_fa.py

[ROCm] Support for Whisper v1 with Aiter Unified Attention and Aiter Flash Attention (#28376)

2025-11-24 03:26:00 +00:00

rocm_aiter_unified_attn.py

[ROCm] Support for Whisper v1 with Aiter Unified Attention and Aiter Flash Attention (#28376)

2025-11-24 03:26:00 +00:00

rocm_attn.py

[ROCm] Support for Whisper v1 with Aiter Unified Attention and Aiter Flash Attention (#28376)

2025-11-24 03:26:00 +00:00

short_conv_attn.py

Remove deprecated fields from CompilationConfig (#27593)

2025-11-12 16:10:28 +00:00

tree_attn.py

[Attention] Refactor FA block_size limitations to hybrid models only (#29084)

2025-11-22 06:38:44 -08:00

triton_attn.py

[Attention] Refactor FA block_size limitations to hybrid models only (#29084)

2025-11-22 06:38:44 -08:00

utils.py

[Feature] Prefill Context Parallel (PCP) basic support (#28718)

2025-11-19 15:52:44 -05:00