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

remove qwen2.py llama.py fix llama output

Browse Source
This commit is contained in:
hanhaowen
2025-12-31 11:31:26 +08:00
parent b3c30a3cb9
commit b015bb76fd
11 changed files with 65 additions and 1263 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
vllm_kunlun/models/qwen3_vl.py
View File

@@ -80,8 +80,8 @@ from .qwen2_5_vl import (Qwen2_5_VisionAttention,
Qwen2_5_VLImagePixelInputs,
Qwen2_5_VLVideoEmbeddingInputs, Qwen2_5_VLVideoInputs,
Qwen2_5_VLVideoPixelInputs)
from .qwen2_vl import Qwen2VLProcessingInfo
from .qwen3 import Qwen3ForCausalLM, Qwen3Model
from vllm.model_executor.models.qwen2_vl import Qwen2VLProcessingInfo
from vllm.model_executor.models.qwen3 import Qwen3ForCausalLM, Qwen3Model
from vllm.model_executor.models.utils import (AutoWeightsLoader, PPMissingLayer, WeightsMapper,
maybe_prefix, merge_multimodal_embeddings)
from vllm.model_executor.models.vision import get_vit_attn_backend, run_dp_sharded_mrope_vision_model