vanshbhatia-amdandvanshbhatia-amd 7ffb98e248 [ROCm] Retune MI355 selective_state_update float32 config on the unified effective_batch grid (#48373) ...

Signed-off-by: vanshbhatia-amd <210711135+vanshbhatia-amd@users.noreply.github.com>
Co-authored-by: vanshbhatia-amd <210711135+vanshbhatia-amd@users.noreply.github.com>

2026-07-14 18:26:35 +00:00

..

gdn

[Perf][Qwen] Replace MOE all-reduce with reduce-scatter (#47006)

2026-07-12 06:14:49 +00:00

linear

[Feature] Support MTP speculative decoding for Bailing hybrid models (#44880)

2026-07-06 10:38:50 +08:00

ops

[ROCm] Retune MI355 selective_state_update float32 config on the unified effective_batch grid (#48373)

2026-07-14 18:26:35 +00:00

__init__.py

[Kernel/Model] Migrate mamba_ssm and causal_conv1d kernels to vLLM (#7651)

2024-08-28 15:06:52 -07:00

abstract.py

DCP supports hybrid attention (#40996)

2026-07-09 21:34:45 -07:00

mamba_mixer2.py

[Bugfix][Hybrid][NemotronH] Fix mamba_cache_mode=all + speculative decoding crash (#41233)

2026-05-18 14:54:00 +03:00

mamba_mixer.py

[Perf][1/n] Eliminate various GPU<->CPU syncs (#41429)

2026-05-11 20:36:03 +00:00

mamba_utils.py

[Kernel] Support DS Mamba tail copy for MTP align mode (#45473)

2026-06-16 22:50:30 +00:00

short_conv.py

feat(cpu): add CPU support for Mamba ShortConv (#35059)

2026-07-07 10:47:08 +08:00