EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

upgrade vLLM to main (#4608)

Browse Source 
1. fix https://github.com/vllm-project/vllm/pull/28542
The model structure modifications we involved in are:
     - Qwen2.5-VL(still exist some patch)
     - Qwen2-VL
     - Qwen2
     - DeepSeek series
     - Qwen-moe series
2. fix https://github.com/vllm-project/vllm/pull/29121
   the output token now  type changed from np to `list[list[int]]`

3. fix https://github.com/vllm-project/vllm/pull/29262
    `xformers` backend for multimodal now has been deprecated
4. fix https://github.com/vllm-project/vllm/pull/29342

5. fix https://github.com/vllm-project/vllm/pull/28579
6. fix https://github.com/vllm-project/vllm/pull/28718
7. fix https://github.com/vllm-project/vllm/issues/28665
8. fix https://github.com/vllm-project/vllm/pull/26847
vllm introduced the `optimization-level`, some default config has been
changed, and the param `--enforce-eager` has been deprecated
9. fix http://github.com/vllm-project/vllm/pull/29223 it retuns tuple
for sampler.
10. fix https://github.com/vllm-project/vllm/pull/29471 we'll remove the
related patch to avoid this kind of error.

Co-authored-by: hfadzxy <starmoon_zhang@163.com>
Co-authored-by: wangli <wangli858794774@gmail.com>


- vLLM version: v0.11.2

---------

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
Signed-off-by: wangli <wangli858794774@gmail.com>
Signed-off-by: hfadzxy <starmoon_zhang@163.com>
Co-authored-by: wangli <wangli858794774@gmail.com>
Co-authored-by: hfadzxy <starmoon_zhang@163.com>
This commit is contained in:
wangxiyuan
2025-12-02 22:10:52 +08:00
committed by GitHub
parent 4588cdac02
commit 7f2673ea2d
60 changed files with 383 additions and 374 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
vllm_ascend/patch/worker/patch_qwen3_vl.py
View File

@@ -23,7 +23,6 @@ import torch.nn as nn
from transformers.models.qwen3_vl.configuration_qwen3_vl import \
Qwen3VLVisionConfig
from vllm.attention.backends.registry import AttentionBackendEnum
from vllm.attention.layer import check_upstream_fa_availability
from vllm.model_executor.layers.activation import _ACTIVATION_REGISTRY
from vllm.model_executor.layers.quantization import QuantizationConfig
from vllm.model_executor.layers.rotary_embedding import get_rope
@@ -101,7 +100,6 @@ class AscendQwen3_VisionTransformer(nn.Module):
head_size=head_dim,
rotary_dim=head_dim // 2,
max_position=8192,
base=10000.0,
is_neox_style=True,
)
@@ -133,17 +131,10 @@ class AscendQwen3_VisionTransformer(nn.Module):
dtype=torch.get_default_dtype(),
attn_backend_override=attn_backend_override,
)
use_upstream_fa = False
if (self.attn_backend != AttentionBackendEnum.FLASH_ATTN
and self.attn_backend != AttentionBackendEnum.ROCM_AITER_FA
and check_upstream_fa_availability(torch.get_default_dtype())):
self.attn_backend = AttentionBackendEnum.FLASH_ATTN
use_upstream_fa = True
if self.attn_backend not in {
AttentionBackendEnum.FLASH_ATTN,
AttentionBackendEnum.TORCH_SDPA,
AttentionBackendEnum.XFORMERS,
AttentionBackendEnum.ROCM_AITER_FA,
}:
raise RuntimeError(
@@ -159,7 +150,6 @@ class AscendQwen3_VisionTransformer(nn.Module):
prefix=f"{prefix}.blocks.{layer_idx}",
use_data_parallel=use_data_parallel,
attn_backend=self.attn_backend,
use_upstream_fa=use_upstream_fa,
) for layer_idx in range(vision_config.depth)
])