dsxsteven 9a885d08d0 [Feat] Multi-stream for eplb heat collection and aggregation (#4214) ...

### What this PR does / why we need it?
This PR optimizes multistream for eplb heat collection and aggregation

- vLLM version: v0.12.0
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.12.0

---------

Signed-off-by: daishixun <dsxsteven@sina.com>
Co-authored-by: Mengqing Cao <cmq0113@163.com>

2025-12-09 16:16:55 +08:00

..

__init__.py

[Refactor] [MoE] Rename moe-related classes & files (#3646)

2025-10-25 11:22:03 +08:00

comm_utils.py

[Refactor] [MoE] Rename moe-related classes & files (#3646)

2025-10-25 11:22:03 +08:00

experts_selector.py

【fix】ops gatingtopk fix nightly ci error (#4340)

2025-12-04 20:09:21 +08:00

fused_moe.py

[Feat] Multi-stream for eplb heat collection and aggregation (#4214)

2025-12-09 16:16:55 +08:00

moe_comm_method.py

add dispatch_gmm_combine kernel (#3532)

2025-12-04 23:00:59 +08:00

moe_mlp.py

[Bugifx] fix quant_apply_mlp w1_scale type error & fix getting num_local_expert (#4632)

2025-12-05 16:04:24 +08:00

prepare_finalize.py

[long_seq] remove long_seq env (#4660)

2025-12-05 10:31:49 +08:00

token_dispatcher.py

add dispatch_gmm_combine kernel (#3532)

2025-12-04 23:00:59 +08:00