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

..

expert_map.json

Add unit test local cpu guide and enable base testcase (#1566)

2025-07-06 10:42:27 +08:00

test_activation.py

[refact] unified soc_version code (#4359)

2025-11-26 14:28:55 +08:00

test_comm_utils.py

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

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

test_expert_load_balancer.py

eplb redundant expert bugfix (#4291)

2025-11-21 14:24:35 +08:00

test_fused_moe.py

[Bugfix] bugfix for moe_mlp (#4822)

2025-12-12 14:51:20 +08:00

test_layernorm.py

Adopt inductor fusion and define quantization fusion pass (#4168)

2025-12-04 10:29:48 +08:00

test_linear.py

[Feat] Add custom Embedding tensor model parallel (#2616)

2025-12-12 14:41:20 +08:00

test_mla.py

Remove ascend schuduler ut (#4684)

2025-12-04 14:10:28 +08:00

test_moe_comm_method.py

[EPLB][Ops] Integerate grouped_matmul_swiglu_quant_weight_nz_tensor_list operator into dynamic EPLB (#4216)

2025-11-30 22:52:05 +08:00

test_prepare_finalize.py

[Perf]enable prefill flashcommon3 (#4065)

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

test_rotary_embedding.py

Drop torchair (#4814)

2025-12-10 09:20:40 +08:00

test_token_dispatcher.py

[perf] replace all_reduce for kv_consumer and support different num_tokens among all ranks (#4983)

2025-12-13 18:59:54 +08:00

test_vocab_parallel_embedding.py

[Feat] Add custom Embedding tensor model parallel (#2616)

2025-12-12 14:41:20 +08:00