Update app.py

app.py

@@ -7,7 +7,7 @@ import streamlit as st
 from llama_index.core import Settings
 from llama_index.llms.cerebras import Cerebras
 from llama_index.core import PromptTemplate
-from llama_index.embeddings.huggingface import HuggingFaceEmbedding
+from llama_index.embeddings.mixedbreadai import MixedbreadAIEmbedding, EncodingFormat
 from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
 from llama_index.readers.docling import DoclingReader
 from llama_index.core.node_parser import MarkdownNodeParser
@@ -68,9 +68,12 @@ with st.sidebar:
 
                     docs = loader.load_data()
 
-                    # setup llm & embedding model
+                    # Setup llm & embedding model
                     llm = load_llm()
-                    embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-large-en-v1.5", trust_remote_code=True)
+                    # Use Mixedbread AI Embedding instead of HuggingFaceEmbedding
+                    mixedbread_api_key = os.getenv("MXBAI_API_KEY")
+                    embed_model = MixedbreadAIEmbedding(api_key=mixedbread_api_key, model_name="mixedbread-ai/mxbai-embed-large-v1")
+
                     # Creating an index over loaded data
                     Settings.embed_model = embed_model
                     node_parser = MarkdownNodeParser()
@@ -111,7 +114,7 @@ with st.sidebar:
 col1, col2 = st.columns([6, 1])
 
 with col1:
-    st.header(f"RAG over Excel using Dockling 🐥 &  Llama-3.3 70B")
+    st.header(f"RAG over Excel using Dockling 🐥 & Llama-3.3 70B")
 
 with col2:
     st.button("Clear ↺", on_click=reset_chat)
@@ -145,10 +148,7 @@ if prompt := st.chat_input("What's up?"):
             full_response += chunk
             message_placeholder.markdown(full_response + "▌")
 
-        # full_response = query_engine.query(prompt)
-
         message_placeholder.markdown(full_response)
-        # st.session_state.context = ctx
 
     # Add assistant response to chat history
-    st.session_state.messages.append({"role": "assistant", "content": full_response})
+    st.session_state.messages.append({"role": "assistant", "content": full_response})