as llm, use gpt-4o-mini

src/rag.py

@@ -22,8 +22,9 @@ class RAG():
         self.k = 3  # Number of relevant chunks to retrieve
 
         # Constants
-        # self.use_model = 'gpt-4o-mini'
-        self.use_model = 'zephyr-7b-beta'
+        self.use_model = 'gpt-4o-mini'
+        # self.use_model = 'zephyr-7b-alpha'
+        # self.use_model = 'zephyr-7b-beta'
         # self.use_model = 'Mistral-Nemo-Base-2407'
 
         # self.use_vectordb = 'chroma'
@@ -124,7 +125,7 @@ class RAG():
             llm = ChatOpenAI(
                 model_name="gpt-4o-mini",
                 temperature=0)
-        elif self.use_model ==  'zephyr-7b-beta':
+        elif self.use_model in ['zephyr-7b-alpha','zephyr-7b-beta'] :
             provider = "hf-inference"
             print(f'As llm, using HF-Endpint: {self.use_model} through provider: {provider}')
             llm = HuggingFaceEndpoint(