Update main.py

main.py

@@ -64,7 +64,7 @@ async def stream(item: validation):
     user="""
     {prompt}"""
     
-    model = Llama(model_path="./code/"+model['file'], n_ctx=model['nctx'], verbose=False,  n_threads=8)
+    llm = Llama(model_path="./code/"+model['file'], n_ctx=model['nctx'], verbose=False,  n_threads=8)
 
     prompt = f"{prefix}{user.replace('{prompt}', item.prompt)}{suffix}"
     return llm(prompt, max_tokens=max_tokens)