Transformers How to use hyan/qwen_speech_stage1 with Transformers:
# Load model directly
from transformers import AutoProcessor, Qwen2VLForConditionalGenerationWithAudio
processor = AutoProcessor.from_pretrained("hyan/qwen_speech_stage1")
model = Qwen2VLForConditionalGenerationWithAudio.from_pretrained("hyan/qwen_speech_stage1", device_map="auto") 
