Jade Zheng 0dfdfa9526 [Feature] Enhance all-reduce skipping logic for MoE models in NPUModelRunner (#5329) ...

Besides enabling `recompute_scheduler_enable`, we can skip all_reduce
when max_num_batched_tokens is below mc2's requirement.

- vLLM version: release/v0.13.0
- vLLM main:
bc0a5a0c08

---------

Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>

2025-12-26 17:39:44 +08:00

..

_cann_ops_custom

[Kernel] add custom op GmmSwigluQuantWeightNzTensorList (#3804)

2025-11-28 18:06:39 +08:00

attention

[Bugfix] Fix Qwen P/D Disaggregation accuracy issue (#5340)

2025-12-25 22:46:08 +08:00

compilation

[feature] support pcp + mtp in full graph (#4572)

2025-12-22 16:13:39 +08:00

core

[CI] fix lint (#5216)

2025-12-20 17:03:25 +08:00

device_allocator

[Misc]Clean up useless import from vllm (#2049)

2025-07-28 16:01:59 +08:00

distributed

[Bugfix] Use hf_text_config instead of hf_config to support multimodal PD-Disaggregated (#5205)

2025-12-22 20:21:45 +08:00

eplb

[Misc] Cleanup useless print and logger (#5220)

2025-12-22 11:28:26 +08:00

kv_offload

upgrade vLLM to main (#4608)

2025-12-02 22:10:52 +08:00

lora

[BugFix]Fix precision issue for LoRA feature (#4141)

2025-12-19 14:22:06 +08:00

model_loader

[CI] speed up ut (#4901)

2025-12-11 18:45:43 +08:00

ops

move contiguous in fused_sigmoid_gating_delta_rule_update to model_runner_v1 (#5274)

2025-12-26 09:19:47 +08:00

patch

[BugFix][Fusion] Patch compile backend to make fusion available (#5308)

2025-12-26 09:18:16 +08:00

quantization

[quantization] Add w8a16 quantization support (#4541)

2025-12-24 19:49:32 +08:00

sample

Revert "Add MagicMTP(block verify) and Triton optimization (#4443)" (#5380)

2025-12-26 15:06:13 +08:00

spec_decode

[main][Refactor] Remove with_prefill parameter from set_ascend_forward_context (#5094)

2025-12-23 14:30:50 +08:00

worker

[Feature] Enhance all-reduce skipping logic for MoE models in NPUModelRunner (#5329)

2025-12-26 17:39:44 +08:00

xlite

[CI] add xlite e2e test (#5305)

2025-12-25 09:17:06 +08:00

__init__.py

clean up model module (#4611)

2025-12-02 17:35:47 +08:00

ascend_config.py

cleanup ascend config (#5296)

2025-12-26 14:07:37 +08:00

ascend_forward_context.py

[Bugfix] Fix unsuitable moe_comm_type under ep=1 scenario (#5388)

2025-12-26 16:45:45 +08:00

cpu_binding.py

[main] support cpu binding (#3546)

2025-10-21 09:17:03 +08:00

envs.py

Remove VLLM_ASCEND_ENABLE_DENSE_OPTIMIZE (#5272)

2025-12-25 11:09:56 +08:00

flash_common3_context.py

[Perf]enable prefill flashcommon3 (#4065)

2025-12-14 09:34:13 +08:00

meta_registration.py

Fix the bugs about operator registration by PyTorch Dispatcher (#2786)

2025-09-13 11:58:52 +08:00

platform.py

cleanup ascend config (#5296)

2025-12-26 14:07:37 +08:00

profiling_config.py

Drop ascend scheduler (#4623)

2025-12-05 09:03:45 +08:00

utils.py

cleanup ascend config (#5296)

2025-12-26 14:07:37 +08:00