周周周 d957ccd46d seq_lens related tensor shape -> [max_num_seqs] (#6535)

2026-03-02 11:18:30 +08:00

..

__init__.py

…

append_attention.py

[Feature]Supports SWA based on appendattn (#6547)

2026-03-01 19:02:08 +08:00

flash_mask_attention.py

seq_lens related tensor shape -> [max_num_seqs] (#6535)

2026-03-02 11:18:30 +08:00

get_attn_mask_q.py

…

get_block_shape_and_split_kv_block.py

…

gqa_rope_write_cache.py

…

init_kv_signal_per_query.py

…

init_signal_layerwise.py

…

open_shm_and_get_meta_signal.py

…

pre_cache_len_concat.py

…