Matthew BonanniandClaude Opus 5.5 624a5ddafd [Attention] Fetch MLA prefill workspace at use instead of init (#60654) ...

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-13 17:25:03 +00:00

..

mla

[Attention] Fetch MLA prefill workspace at use instead of init (#60654)

2026-10-13 17:25:03 +00:00

__init__.py

…

b12x.py

[Kernel] Resubmit PR 48666 - Gemma4 FP8 KV FA4 head dim 512 backend selection (#53175)

2026-09-23 20:29:06 -07:00

composite.py

[KV Cache] GLM-5.3-Flash: use the generic packed KV layout; CircularBufferSpec tail with speculative ring slots (#57169)

2026-10-09 15:48:37 -04:00

cpu_attn.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

fa_utils.py

[Kernel] Resubmit PR 48666 - Gemma4 FP8 KV FA4 head dim 512 backend selection (#53175)

2026-09-23 20:29:06 -07:00

flash_attn_diffkv.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

flash_attn.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

flashinfer.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

flex_attention.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

gdn_attn.py

[ROCm][Perf] Add AITER FlyDSL GDN prefill backend (#57560)

2026-10-07 09:44:22 -05:00

hpc_attn.py

[Kernel] Resubmit PR 48666 - Gemma4 FP8 KV FA4 head dim 512 backend selection (#53175)

2026-09-23 20:29:06 -07:00

linear_attn.py

[Attention]: Use KVCacheSpec for AttentionMetadataBuilder type hints (#50148)

2026-07-31 11:20:18 +00:00

mamba1_attn.py

[Core] Remove deprecated mamba_cache_mode "all" (#58997)

2026-10-01 02:59:37 -07:00

mamba2_attn.py

[Feat][Mamba2] Enable internal prefill checkpoints (#57329)

2026-10-06 07:53:51 +00:00

mamba_attn.py

[Core][Mamba] Remove leftovers of mamba_cache_mode="all" (#60043)

2026-10-13 00:58:59 -07:00

recoverssm_metadata.py

[K3] support recoverssm for K3 (#51855)

2026-08-17 19:34:14 +08:00

registry.py

[Attention] Enable FA4 sparse MQA decode on Blackwell (#55866)

2026-10-13 09:31:37 -04:00

rocm_aiter_fa.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

rocm_aiter_unified_attn.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

rocm_attn.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

rocm_segmented_attn.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

short_conv_attn.py

[Core] Remove deprecated mamba_cache_mode "all" (#58997)

2026-10-01 02:59:37 -07:00

triton_attn_diffkv.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

triton_attn.py

[Attention][Spec Decode] Multi-query Triton split-K for spec-decode verify, with empty-segment/padding hardening (#56148)

2026-10-13 04:37:56 +04:00

triton_flash_attn.py

[Kernel] Resubmit PR 48666 - Gemma4 FP8 KV FA4 head dim 512 backend selection (#53175)

2026-09-23 20:29:06 -07:00

triton_flashinfer.py

[Attention] Add Triton/FlashInfer composite for multimodal prefix attention (#56305)

2026-09-13 14:29:41 +00:00

turboquant_attn.py

[Kernel] Resubmit PR 48666 - Gemma4 FP8 KV FA4 head dim 512 backend selection (#53175)

2026-09-23 20:29:06 -07:00

ultraquant_attn.py

[Attention] Add UltraQuant 4-bit KV cache backend (FlyDSL D=256) (#57057)

2026-10-06 20:42:52 +00:00

utils.py

[Model][GLM-5.3-Flash] Support DCP for the kpool sparse indexer (#59211)

2026-10-05 16:44:14 +02:00

zentorch_sdpa.py

[Attention][CPU] Run Zen CPU encoder attention on zentorch SDPA (#54508)

2026-09-24 02:05:44 -07:00