AlvisGong ba28d54f35 [Perf]enable prefill flashcommon3 (#4065) ...

### What this PR does / why we need it?
moe multistream overlap to improve the performance.

### How was this patch tested?
--additional-config '{"multistream_overlap_gate": true}'

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: AlvisGong <gwly0401@163.com>
Signed-off-by: chenxiao <Jaychou1620@Gmail.com>
Co-authored-by: clrs97 <524936896@qq.com>
Co-authored-by: zzhx1 <zzh_201018@outlook.com>
Co-authored-by: chenxiao <Jaychou1620@Gmail.com>

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

..

cpu_offload_manager

upgrade vLLM to main (#4608)

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

device_communicators

[MISC] Clean up torch_npu (#688)

2025-04-29 18:03:38 +08:00

kvpool

[usability]Modify the default value of the protocol to ascend (#4959)

2025-12-12 16:56:18 +08:00

__init__.py

[P/D][main]Offline the llmdatadist connector related parts of the code and files. (#4780)

2025-12-09 22:36:43 +08:00

communicator.py

[2/N][Feat] Add MC2 communication method for MoE layers (#2469)

2025-08-26 19:05:23 +08:00

cpu_offload_connector.py

upgrade vLLM to main (#4608)

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

mooncake_connector.py

Revert "[Bugfix] support mtp kv transfer and pp partition by hand in kv transfer (#4892)" (#4981)

2025-12-13 18:58:55 +08:00

mooncake_layerwise_connector.py

[P/D] check kv extra config and del hccl backend (#4547)

2025-12-07 15:19:42 +08:00

mooncake_transfer_engine.py

[Feature][main]reconstruction kvpool connector to ascend connector (#4438)

2025-11-28 18:08:37 +08:00

parallel_state.py

[Perf]enable prefill flashcommon3 (#4065)

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

utils.py

[Perf]enable prefill flashcommon3 (#4065)

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