tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

fix: qwen3_vl attention config (#3216)

Browse Source
This commit is contained in:
NanoCode012
2025-10-17 10:35:03 +07:00
committed by GitHub
parent aa1240acd8
commit 93ba57396f
1 changed files with 5 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
src/axolotl/monkeypatch/lora_kernels.py
View File

@@ -134,6 +134,11 @@ def get_attention_cls_from_config(cfg: DictDefault) -> Type[nn.Module]:
return Qwen2Attention return Qwen2Attention
if model_type == "qwen3_vl":
from transformers.models.qwen3_vl.modeling_qwen3_vl import Qwen3VLTextAttention
return Qwen3VLTextAttention
if model_type == "mllama": if model_type == "mllama":
from transformers.models.mllama.modeling_mllama import MllamaTextSelfAttention from transformers.models.mllama.modeling_mllama import MllamaTextSelfAttention