Update app.py

app.py

@@ -8,14 +8,14 @@ import os
 
 MODEL_NAME = "openai/whisper-large-v3-turbo"
 BATCH_SIZE = 8
-FILE_LIMIT_MB = 1000
+FILE_LIMIT_MB = 5000
 MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 
 device = 0 if torch.cuda.is_available() else "cpu"
 
 # Initialize the LLM
 if torch.cuda.is_available():
-    llm_model_id = "NousResearch/Meta-Llama-3.1-8B-Instruct"
+    llm_model_id = "chuanli11/Llama-3.2-3B-Instruct-uncensored"
     llm = AutoModelForCausalLM.from_pretrained(llm_model_id, torch_dtype=torch.float16, device_map="auto")
     tokenizer = AutoTokenizer.from_pretrained(llm_model_id)
     tokenizer.use_default_system_prompt = False