zouyida2052 05a471001b bugfix for qwen2_5_vl (#805) ...

### What this PR does / why we need it?
the interface of qwen2.5vl changes from column linear to qkv linear,
this makes our weight pad func become abnormal, thus we optimize
split_qkv func to fix this bug.

### Does this PR introduce _any_ user-facing change?
no

### How was this patch tested?
with CI

Signed-off-by: zouyida2052 <zouyida2002@gmail.com>

2025-05-29 17:20:39 +08:00

..

attention

[Bugfix][V1] Fix deepseek with v1 (#958)

2025-05-29 11:57:43 +08:00

compilation

[aclgraph] implentment NPUPiecewiseBackend to enable aclgraph (#836)

2025-05-29 11:58:26 +08:00

core

[Feature] Impl v1 disaggregated prefill in ascend scheduler (#852)

2025-05-23 10:15:29 +08:00

device_allocator

[bugfix] Improve log level and info for custom ops build (#937)

2025-05-23 10:05:57 +08:00

distributed

[CI/UT][PD Disaggreate] Initialize PD Disaggreate UT (#889)

2025-05-29 10:17:12 +08:00

lora

[Bugfix] fix import error (#600)

2025-04-22 08:57:25 +08:00

models

bugfix for qwen2_5_vl (#805)

2025-05-29 17:20:39 +08:00

ops

[aclgraph] implentment NPUPiecewiseBackend to enable aclgraph (#836)

2025-05-29 11:58:26 +08:00

patch

[CI] upgrade to vllm 0.9.0 (#959)

2025-05-28 21:18:41 +08:00

quantization

[Perf] Refactor tensor disposal logic to reduce memory usage (#966)

2025-05-29 11:48:26 +08:00

sample

Spec decode support for V1 Engine (#874)

2025-05-23 14:25:46 +08:00

worker

[Perf] Refactor tensor disposal logic to reduce memory usage (#966)

2025-05-29 11:48:26 +08:00

__init__.py

[CI] Patch torch.library.infer_schema for fused moe ops to fix CI (#854)

2025-05-14 19:49:09 +08:00

envs.py

[Bugfix] Fix the method of importing environment variables in DeepSee… (#817)

2025-05-13 12:52:30 +08:00

platform.py

[aclgraph] implentment NPUPiecewiseBackend to enable aclgraph (#836)

2025-05-29 11:58:26 +08:00

utils.py

[Perf] Refactor tensor disposal logic to reduce memory usage (#966)

2025-05-29 11:48:26 +08:00