Implemented Groq

app02-chatRag.py

@@ -17,6 +17,9 @@ from langchain_community.embeddings import HuggingFaceEmbeddings
 # GeminiPro
 from langchain_google_genai import ChatGoogleGenerativeAI
 
+# Groq
+from langchain_groq import ChatGroq
+
 # Pinecone vector database
 from pinecone import Pinecone, ServerlessSpec
 from langchain_pinecone import PineconeVectorStore
@@ -26,10 +29,18 @@ setid = "global"
 
 embeddings = HuggingFaceEmbeddings(model_name=os.getenv("EMBEDDINGS_MODEL"))
 
+# OpenAI
 # model = ChatOpenAI(temperature=0.0)
-model = ChatGoogleGenerativeAI(
-    model="gemini-pro", temperature=0.1, convert_system_message_to_human=True
-)
+
+# Gemini
+# model = ChatGoogleGenerativeAI(
+#     model="gemini-pro", temperature=0.1, convert_system_message_to_human=True
+# )
+
+# Groq
+# llama2-70b-4096 (4k), mixtral-8x7b-32768 (32k)
+model = ChatGroq(model_name='mixtral-8x7b-32768')
+
 
 pc = Pinecone(
         api_key=os.getenv("PINECONE_API_KEY")
@@ -81,7 +92,7 @@ def rag_query(question: str, history: list[list[str]]):
             | pipeLog
             | model
         )
-        response = chain.invoke({ "chat_history": chat_history, "question": "Que es consenso?"})
+        response = chain.invoke({ "chat_history": chat_history, "question": question })
         return response.content
 
 

modules.md

@@ -1,5 +1,7 @@
 ```sh
 
+pip install ipykernel IProgress ipywidgets --upgrade
+
 pip install \
     gradio                          \
     langchain langchain-community   \
@@ -8,7 +10,11 @@ pip install \
     langchain-pinecone              \
     huggingface_hub
 
-pip install ipykernel IProgress ipywidgets --upgrade
+# Groq API
+pip install langchain_groq
+
+
+
 
     # python-dotenv                   \
     # pinecone-client==2.2.4          \