Default model

inference.py

@@ -12,8 +12,6 @@ llm = Llama.from_pretrained(
 
 @app.route("/v1/completions", methods=["POST"])
 def generate():
-    mem_info = subprocess.run(["cat", "b/proc/meminfo", "|", "grep", "MemTotal"], capture_output=True)
-    print (mem_info)
     data = request.json
     prompt = data.get("prompt", "")
     max_tokens = data.get("max_tokens", 50)