Update backend/query_llm.py

backend/query_llm.py

@@ -18,10 +18,10 @@ repetition_penalty = 1.2
 OPENAI_KEY = getenv("OPENAI_API_KEY")
 HF_TOKEN = getenv("HUGGING_FACE_HUB_TOKEN")
 
-hf_client = InferenceClient(
-       "mistralai/Mistral-7B-Instruct-v0.1",
-       token=HF_TOKEN
-       )
+# hf_client = InferenceClient(
+#        "mistralai/Mistral-7B-Instruct-v0.1",
+#        token=HF_TOKEN
+#        )
 
 
 # hf_client = InferenceClient(
@@ -87,7 +87,7 @@ def generate_hf(prompt: str, history: str, temperature: float = 0.5, max_new_tok
         #                                     stream=True, details=True, return_full_text=False)
         client_Qwen = Client("Qwen/Qwen1.5-110B-Chat-demo")
         response = client_Qwen.predict(
-                    query=formatted_prompt,
+                    query=prompt,
                     history=[],
                     system="You are a helpful assistant.",
                     api_name="/model_chat"