sychen52 947138b6c2 Add nvfp4 kv cache support (#40177) ...

Signed-off-by: Shiyang Chen <shiychen@nvidia.com>

2026-05-01 04:55:16 +00:00

..

mla

[Bugfix][MLA] Size arange_buffer to max_num_batched_tokens to prevent CUDA IMA (#39277)

2026-04-29 16:14:50 -07:00

__init__.py

[V1] Implement vLLM V1 [1/N] (#9289)

2024-10-22 01:24:07 -07:00

cpu_attn.py

[Feat] CPU fp8 attn for AMX/AVX-512 (#39445)

2026-04-29 20:43:21 +08:00

fa_utils.py

[Attention] use diff kv backend for mimo v2 flash (#40045)

2026-04-24 11:25:55 +00:00

flash_attn_diffkv.py

[Attention] use diff kv backend for mimo v2 flash (#40045)

2026-04-24 11:25:55 +00:00

flash_attn.py

[Bugfix] Make Attention Backend Auto-Selection Batch-Invariance-Aware (#40193)

2026-04-23 14:57:03 +00:00

flashinfer.py

Add nvfp4 kv cache support (#40177)

2026-05-01 04:55:16 +00:00

flex_attention.py

(bugfix): block_size check for flex attn (#41363)

2026-04-30 18:59:26 -07:00

gdn_attn.py

[Models][GDN] Remove GPU/CPU syncs in GDNAttentionMetadata.build during speculative decoding (#38047)

2026-04-06 15:39:37 +00:00

linear_attn.py

[HMA]Fix corner case when hybrid page_size can not be evenly divided issue (blk_size=64,tp=4) (#37467)

2026-03-30 16:47:30 +00:00

mamba1_attn.py

[HMA]Fix corner case when hybrid page_size can not be evenly divided issue (blk_size=64,tp=4) (#37467)

2026-03-30 16:47:30 +00:00

mamba2_attn.py

[HMA]Fix corner case when hybrid page_size can not be evenly divided issue (blk_size=64,tp=4) (#37467)

2026-03-30 16:47:30 +00:00

mamba_attn.py

[Bugfix] Use null block (0) for padded block table entries (#35431)

2026-03-30 14:02:51 -07:00

registry.py

[Attention Backend] TurboQuant: 2-bit KV cache compression with 4x capacity (#38479)

2026-04-14 19:57:13 -07:00

rocm_aiter_fa.py

[Feat] dflash support for ROCm (#39703)

2026-04-21 14:58:20 +08:00

rocm_aiter_unified_attn.py

[ROCm] Support non-causal attention in ROCM_ATTN (#40176)

2026-04-22 12:57:12 +09:00

rocm_attn.py

[ROCm] Support non-causal attention in ROCM_ATTN (#40176)

2026-04-22 12:57:12 +09:00

short_conv_attn.py

[HMA]Fix corner case when hybrid page_size can not be evenly divided issue (blk_size=64,tp=4) (#37467)

2026-03-30 16:47:30 +00:00

tree_attn.py

[Attention][V0 Deprecation] Deprecate accept output buffer (#39125)

2026-04-07 17:14:58 -04:00

triton_attn.py

[Bugfix] Make Attention Backend Auto-Selection Batch-Invariance-Aware (#40193)

2026-04-23 14:57:03 +00:00

turboquant_attn.py

[Attention][TurboQuant] Share dequant buffers, eliminate float16_copy (#40941)

2026-04-27 13:48:36 +08:00

utils.py

[Feat] DeepSeek V4 Rebased (#40860)

2026-04-26 18:31:08 -07:00