Update app.py

app.py

@@ -139,6 +139,7 @@ async def create_upload_file(text_field: str, file: UploadFile = File(...), ocr:
     indexing.add_component("sparse_doc_embedder", FastembedSparseDocumentEmbedder(model="facebook/spar-wiki-bm25-lexmodel-context-encoder"))
     indexing.add_component("dense_doc_embedder", FastembedDocumentEmbedder(model="sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2"))
     indexing.add_component("writer", DocumentWriter(document_store=document_store, policy=DuplicatePolicy.OVERWRITE))
+    
     indexing.connect("sparse_doc_embedder", "dense_doc_embedder")
     indexing.connect("dense_doc_embedder", "writer")
     
@@ -161,11 +162,20 @@ def search(prompt: str):
     )
     from haystack.components.rankers import TransformersSimilarityRanker
     from haystack.components.joiners import DocumentJoiner
-
+    from haystack.components.generators import OpenAIGenerator
+    from haystack.utils import Secret
+    
     start_time = time.time()
     
     # Querying
 
+
+    generator = OpenAIGenerator(
+        api_key=Secret.from_env_var("OCTOAI_TOKEN"),
+        api_base_url="https://text.octoai.run/v1",
+        model="meta-llama-3-8b-instruct",
+        generation_kwargs = {"max_tokens": 512}
+    )
     querying = Pipeline()
     querying.add_component("sparse_text_embedder", FastembedSparseTextEmbedder(model="facebook/spar-wiki-bm25-lexmodel-query-encoder'"))
     querying.add_component("dense_text_embedder", FastembedTextEmbedder(
@@ -174,15 +184,20 @@ def search(prompt: str):
     querying.add_component("retriever", QdrantHybridRetriever(document_store=document_store))
     querying.add_component("document_joiner", DocumentJoiner())
     querying.add_component("ranker", TransformersSimilarityRanker(model="BAAI/bge-m3"))
+    querying.add_component("llm", generator)
+    
     querying.connect("sparse_text_embedder.sparse_embedding", "retriever.query_sparse_embedding")
     querying.connect("dense_text_embedder.embedding", "retriever.query_embedding")
     querying.connect("retriever", "document_joiner")
     querying.connect("document_joiner", "ranker")
+    pipeline.connect("ranker", "llm")
     
     results = querying.run(
         {"dense_text_embedder": {"text": prompt},
          "sparse_text_embedder": {"text": prompt},
-         "ranker": {"query": prompt}}
+         "ranker": {"query": prompt},
+         "llm": {"query": prompt},
+        }
     )