init

app.py

@@ -17,10 +17,12 @@ model = AutoModelForCausalLM.from_pretrained(
     MODEL_ID,
     device_map="cuda" if torch.cuda.is_available() else "cpu",
     torch_dtype="auto",
-    trust_remote_code=True,
     _attn_implementation="flash_attention_2",
 )
 
+model.load_adapter(MODEL_ID, adapter_name="speech", device_map="cuda" if torch.cuda.is_available() else "cpu", adapter_kwargs={"subfolder": 'speech-lora'})
+model.set_adapter("speech")
+
 generation_config = GenerationConfig.from_pretrained(MODEL_ID)
 
 @spaces.GPU