Fix attention mask and padding token ID warning

llama_models.py

@@ -19,8 +19,12 @@ def load_model(model_name):
  return tokenizer, model
 
 async def process_text_local(model_name, text):
+ print("Loading model and tokenizer...")
  tokenizer, model = load_model(model_name)
- inputs = tokenizer.encode(text, return_tensors="pt")
- outputs = model.generate(inputs, max_length=512)
+ print("Encoding text...")
+ inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True)
+ print("Generating output...")
+ outputs = model.generate(**inputs, max_length=512)
+ print("Decoding output...")
  result = tokenizer.decode(outputs[0], skip_special_tokens=True)
  return result