drslark 8fb0ef5ffa [main][BugFix] Fixed an accuracy bug of Qwen3-next-MTP when batched inferring (#4932) ...

### What this PR does / why we need it?
Fixes an accuracy bug of Qwen3-next-MTP when batched inferring.
It is descibed in
https://github.com/vllm-project/vllm-ascend/issues/4930.

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

Signed-off-by: drslark <slarksblood@qq.com>

2025-12-15 13:22:30 +08:00

..

_cann_ops_custom

[Kernel] add custom op GmmSwigluQuantWeightNzTensorList (#3804)

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

attention

[Refactor]3/N Refactor mla_v1.py & extract mla_cp (#4933)

2025-12-15 12:59:18 +08:00

compilation

[Fusion] normalize fusion naming and enable e2e test (#4693)

2025-12-11 17:53:43 +08:00

core

[bugfix][refactor] fix recompute_scheduler break with vllm 0.12.0 & support async scheduling & refactor recompute_scheduler.py (#4895)

2025-12-11 22:24:49 +08:00

device_allocator

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

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

distributed

[Perf]enable prefill flashcommon3 (#4065)

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

eplb

BugFix: Resolve PolicyFlashlb warm up function attribute error (#4741)

2025-12-12 14:55:26 +08:00

kv_offload

upgrade vLLM to main (#4608)

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

lora

[refact] unified soc_version code (#4359)

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

model_loader

[CI] speed up ut (#4901)

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

ops

[Perf]enable prefill flashcommon3 (#4065)

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

patch

[main][BugFix] Fixed an accuracy bug of Qwen3-next-MTP when batched inferring (#4932)

2025-12-15 13:22:30 +08:00

quantization

[CI] CI refactor (#4928)

2025-12-14 11:09:56 +08:00

sample

[Performance] Pre-issued exponential distribution operator. (#4908)

2025-12-11 23:02:51 +08:00

spec_decode

[Fix] Fixes issues in MTP with async scheduling and ACL graph (#4963)

2025-12-14 00:10:11 +08:00

worker

[CI][Bugfix] Fix scheduleroutput has no attr get error in prompt logprobs (#4998)

2025-12-15 11:10:39 +08:00

xlite

[Feat] Add Euler xlite graph wrapper support (#4526)

2025-12-08 08:27:46 +08:00

__init__.py

clean up model module (#4611)

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

ascend_config.py

[Perf]enable prefill flashcommon3 (#4065)

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

ascend_forward_context.py

[Feature] model_runner refactor (#4764)

2025-12-12 17:27:09 +08:00

cpu_binding.py

[main] support cpu binding (#3546)

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

envs.py

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

2025-12-12 14:41:20 +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

[bugfix][refactor] fix recompute_scheduler break with vllm 0.12.0 & support async scheduling & refactor recompute_scheduler.py (#4895)

2025-12-11 22:24:49 +08:00

profiling_config.py

Drop ascend scheduler (#4623)

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

utils.py

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

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