4 people f51193b9ae [Kernel][Mamba] Fused-kernel support for align-mode DS-conv state migration with num_accepted_tokens > 1 (#49291) ...

Signed-off-by: Sungsoo Ha <sungsooh@nvidia.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>

2026-07-29 18:54:52 +08:00

..

general

[Kernel][Mamba] Fused-kernel support for align-mode DS-conv state migration with num_accepted_tokens > 1 (#49291)

2026-07-29 18:54:52 +08:00

spec_decode

[ROCm][CI] Stabilize ngram and suffix correctness test (#50190)

2026-07-28 20:29:00 -06:00

__init__.py

[V1] Implement Cascade Attention (#11635)

2025-01-01 21:56:46 +09:00

test_cpu_linear_attn_chunked_prefix.py

[CPU] Enable chunked prefill and prefix caching for qwen3.5 (#46202)

2026-06-25 03:49:21 +00:00

test_hybrid_chunked_prefill.py

[ROCm][CI] Enable hybrid chunked prefill test (#38317)

2026-03-30 10:30:26 +08:00

test_replayssm_decode.py

[Kernel] ReplaySSM: cache SSM inputs for faster Mamba2 standard decode (#48018)

2026-07-24 09:39:49 -07:00