Update app.py

app.py

@@ -58,7 +58,7 @@ def init_vector_search() -> MongoDBAtlasVectorSearch:
     from langchain_community.vectorstores import MongoDBAtlasVectorSearch
 
     HF_TOKEN = os.getenv("HF_TOKEN", "").strip()
-    model_name = "sentence-transformers/paraphrase-MiniLM-L6-v2"
+    model_name = "intfloat/e5-small-v2"
 
     try:
         st.write(f"🔌 Connecting to Hugging Face model: `{model_name}`")
@@ -114,7 +114,7 @@ def generate_response(input_dict: Dict[str, Any]) -> str:
 
     try:
         response = client.chat.completions.create(
-            model="Qwen/Qwen2.5-1.5B-Instruct",
+            model="HuggingFaceH4/zephyr-7b-beta",
             messages=[
                 {"role": "system", "content": prompt},
                 {"role": "user", "content": input_dict["question"]},
@@ -418,13 +418,14 @@ if __name__ == "__main__":
 
 #     #     Notes
 #     #     -----
-#     #     Uses Qwen2.5-1.5B-Instruct model with controlled temperature
+#     #     Uses Zephyr model with controlled temperature
 #     #     """
 #     #     hf_client = InferenceClient(api_key=self.HF_TOKEN)
 #     #     formatted_prompt = self.prompt.format(**input_dict)
 
 #     #     response = hf_client.chat.completions.create(
-#     #         model="Qwen/Qwen2.5-1.5B-Instruct",
+#     #         model="HuggingFaceH4/zephyr-7b-beta"
+,
 #     #         messages=[
 #     #             {"role": "system", "content": formatted_prompt},
 #     #             {"role": "user", "content": input_dict["question"]},