Rag/rag transition 2

lib/chatbot/application.ex

@@ -8,12 +8,6 @@ defmodule Chatbot.Application do
   @impl true
   def start(_type, _args) do
     children = [
-      {Nx.Serving,
-       [
-         serving: Chatbot.Rag.Serving.build_llm_serving(),
-         name: Rag.LLMServing,
-         batch_timeout: 100
-       ]},
       {Nx.Serving,
        [
          serving: Chatbot.Rag.Serving.build_embedding_serving(),

lib/chatbot/rag.ex

@@ -5,7 +5,7 @@
   import Ecto.Query
   import Pgvector.Ecto.Query
 
-  @provider Ai.Nx.new(%{embeddings_serving: Rag.EmbeddingServing, text_serving: Rag.LLMServing})
+  @provider Ai.Nx.new(%{embeddings_serving: Rag.EmbeddingServing})
 
   def ingest(path) do
     path
@@ -46,9 +46,9 @@
     Enum.map(chunks, &Map.put(ingestion, :chunk, &1.text))
   end
 
   def query(query) do
     generation =
       Generation.new(query)
       |> Embedding.generate_embedding(@provider)
       |> Retrieval.retrieve(:fulltext_results, fn generation -> query_fulltext(generation) end)
       |> Retrieval.retrieve(:semantic_results, fn generation ->
@@ -85,7 +85,7 @@
     |> Map.put_new(:updated_at, now)
   end
 
   defp query_with_pgvector(%{query_embedding: query_embedding}, limit \\ 3) do
     {:ok,
      Repo.all(
        from(c in Chatbot.Rag.Chunk,
@@ -95,7 +95,7 @@
      )}
   end
 
   defp query_fulltext(%{query: query}, limit \\ 3) do
     query = query |> String.trim() |> String.replace(" ", " & ")
 
     {:ok,
@@ -107,7 +107,7 @@
      )}
   end
 
   defp smollm_prompt(query, context) do
     """
     <|im_start|>system
     You are a helpful assistant.<|im_end|>