XDaoHong 2d288ba8c9 [Bugfix] fix npu get kv_item_lens in PD separation when use ASCEND_US… (#15852) ...

Co-authored-by: ZhengdQin <zhengdqin@gmail.com>

2026-03-23 15:56:47 +08:00

..

attention

[NPU] [DLLM]DLLM LLaDA2.x graph mode support with NPU speedup modifications (#18485)

2026-03-09 22:41:05 +08:00

graph_runner

[NPU] [DLLM]DLLM LLaDA2.x graph mode support with NPU speedup modifications (#18485)

2026-03-09 22:41:05 +08:00

modules

[NPU][QwenVL] Support qwen image preprocess on npu (#20189)

2026-03-11 15:03:08 +08:00

moe

[NPU] support the Enable return routed experts (#17025)

2026-02-01 18:31:39 +08:00

quantization

[NPU] Support dequant_swiglu_quant & moe_init_routing_v2 & npu_moe_token_unpermute for W8A8 MoE decode (#19913)

2026-03-17 21:39:29 +08:00

allocator_npu.py

Moving _alloc_extend_naive out of npu allocator (#18200)

2026-02-04 02:09:55 -08:00

cmo.py

…

memory_pool_npu.py

[Bugfix] fix npu get kv_item_lens in PD separation when use ASCEND_US… (#15852)

2026-03-23 15:56:47 +08:00

utils.py

[NPU] add new fusion operator DispatchFFNCombine (#20245)

2026-03-18 15:22:04 +08:00