Update app.py

app.py

@@ -28,7 +28,7 @@ from transformers import AutoTokenizer
 
 # Function to calculate token length using Hugging Face tokenizer
 def hf_token_len(text):
-    tokenizer = AutoTokenizer.from_pretrained("Upstage/SOLAR-10.7B-v1.0")
+    tokenizer = AutoTokenizer.from_pretrained("Upstage/SOLAR-1-mini-chat")
     tokens = tokenizer.encode(text)
     return len(tokens)
 
@@ -46,7 +46,7 @@ split_chunks = text_splitter.split_documents(docs)
 # Load the embeddings model
 from langchain.embeddings import HuggingFaceEmbeddings
 
-embedding_model = HuggingFaceEmbeddings(model_name="solar-10.7b")
+embedding_model = HuggingFaceEmbeddings(model_name="solar-embedding-1-large")
 
 # Load the vector store and retriever from Qdrant
 from langchain.vectorstores import Qdrant
@@ -61,8 +61,8 @@ qdrant_vectorstore = Qdrant.from_documents(
 qdrant_retriever = qdrant_vectorstore.as_retriever()
 
 # 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")
+tokenizer = AutoTokenizer.from_pretrained("Upstage/SOLAR-1-mini-chat")
+model = AutoModelForCausalLM.from_pretrained("Upstage/SOLAR-1-mini-chat")
 
 from langchain.prompts import ChatPromptTemplate
 
@@ -128,7 +128,7 @@ retrieval_augmented_qa_chain = (
 @cl.on_chat_start
 async def start_chat():
     settings = {
-        "model": "solar-10.7b",
+        "model": ""Upstage/SOLAR-1-mini-chat",
         "temperature": 0,
         "max_tokens": 500,
         "top_p": 1,