s

app.py

@@ -1,14 +1,14 @@
-# HACK AI   MAKERSPACE PREPR 
+#
+# HACK AI MAKERSPACE PREPR 
 # Date: 2024-5-16
 
 # Basic Imports & Setup
 import os
-from openai import AsyncOpenAI
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
 # Using Chainlit for our UI
 import chainlit as cl
 from chainlit.prompt import Prompt, PromptMessage
-from chainlit.playground.providers import ChatOpenAI
 
 # Getting the API key from the .env file
 from dotenv import load_dotenv
@@ -27,7 +27,7 @@ docs = PyMuPDFLoader(direct_url).load()
 
 import tiktoken
 def tiktoken_len(text):
-    tokens = tiktoken.encoding_for_model("gpt-3.5-turbo").encode(
+    tokens = tiktoken.encoding_for_model("solar-10.7b").encode(
         text,
     )
     return len(tokens)
@@ -44,12 +44,12 @@ text_splitter = RecursiveCharacterTextSplitter(
 split_chunks = text_splitter.split_documents(docs)
 
 # Load the embeddings model
-from langchain_openai.embeddings import OpenAIEmbeddings
+from langchain.embeddings import HuggingFaceEmbeddings
 
-embedding_model = OpenAIEmbeddings(model="text-embedding-3-small")
+embedding_model = HuggingFaceEmbeddings(model_name="solar-10.7b")
 
 # Load the vector store and retriever from Qdrant
-from langchain_community.vectorstores import Qdrant
+from langchain.vectorstores import Qdrant
 
 qdrant_vectorstore = Qdrant.from_documents(
     split_chunks,
@@ -60,10 +60,11 @@ qdrant_vectorstore = Qdrant.from_documents(
 
 qdrant_retriever = qdrant_vectorstore.as_retriever()
 
-from langchain_openai import ChatOpenAI
-openai_chat_model = ChatOpenAI(model="gpt-3.5-turbo")
+# Load the Solar 10.7B model
+tokenizer = AutoTokenizer.from_pretrained("Upstage/SOLAR-10.7B-v1.0")
+model = AutoModelForCausalLM.from_pretrained("Upstage/SOLAR-10.7B-v1.0")
 
-from langchain_core.prompts import ChatPromptTemplate
+from langchain.prompts import ChatPromptTemplate
 
 RAG_PROMPT = """
 SYSTEM:
@@ -120,14 +121,14 @@ from langchain.schema.runnable import RunnablePassthrough
 retrieval_augmented_qa_chain = (
     {"context": itemgetter("question") | qdrant_retriever, "question": itemgetter("question")}
     | RunnablePassthrough.assign(context=itemgetter("context"))
-    | {"response": rag_prompt | openai_chat_model, "context": itemgetter("context")}
+    | {"response": rag_prompt | model, "context": itemgetter("context")}
 )
 
 # Chainlit App
 @cl.on_chat_start
 async def start_chat():
     settings = {
-        "model": "gpt-3.5-turbo",
+        "model": "solar-10.7b",
         "temperature": 0,
         "max_tokens": 500,
         "top_p": 1,
@@ -145,3 +146,5 @@ async def main(message: cl.Message):
 
     msg = cl.Message(content=chainlit_answer)
     await msg.send()
+
+