diff --git a/example-apps/chatbot-rag-app/.flaskenv b/example-apps/chatbot-rag-app/.flaskenv
deleted file mode 100644
index 88612f8d..00000000
--- a/example-apps/chatbot-rag-app/.flaskenv
+++ /dev/null
@@ -1,4 +0,0 @@
-FLASK_APP=api/app.py
-FLASK_RUN_PORT=4000
-# Production mode ensures we don't run into problems.
-FLASK_ENV=production
diff --git a/example-apps/chatbot-rag-app/Dockerfile b/example-apps/chatbot-rag-app/Dockerfile
index 317ebd69..bca2b036 100644
--- a/example-apps/chatbot-rag-app/Dockerfile
+++ b/example-apps/chatbot-rag-app/Dockerfile
@@ -1,4 +1,4 @@
-FROM node:20-alpine AS build-step
+FROM node:22-alpine AS build-step
 WORKDIR /app
 ENV PATH=/node_modules/.bin:$PATH
 COPY frontend ./frontend
@@ -28,7 +28,4 @@ COPY api ./api
 COPY data ./data
 
 EXPOSE 4000
-# The only thing different from running local is that in docker we need to
-# listen on all IPs, not just localhost.
-ENV FLASK_RUN_HOST=0.0.0.0
-CMD [ "flask", "run"]
+CMD [ "python", "api/app.py"]
diff --git a/example-apps/chatbot-rag-app/README.md b/example-apps/chatbot-rag-app/README.md
index 0419fae4..a4b5abfa 100644
--- a/example-apps/chatbot-rag-app/README.md
+++ b/example-apps/chatbot-rag-app/README.md
@@ -67,6 +67,12 @@ docker compose up --build --force-recreate
 *Note*: First time creating the index can fail on timeout. Wait a few minutes
 and retry.
 
+Clean up when finished, like this:
+
+```bash
+docker compose down
+```
+
 ### Run locally
 
 If you want to run this example with Python and Node.js, you need to do a few
@@ -95,9 +101,8 @@ correct packages installed:
 ```bash
 python3 -m venv .venv
 source .venv/bin/activate
-# install dev requirements for pip-compile and dotenv
-pip install pip-tools "python-dotenv[cli]"
-pip-compile
+# Install dotenv which is a portable way to load environment variables.
+pip install "python-dotenv[cli]"
 pip install -r requirements.txt
 ```
 
@@ -105,7 +110,7 @@ pip install -r requirements.txt
 
 First, ingest the data into elasticsearch:
 ```bash
-$ dotenv run -- flask create-index
+$ dotenv run -- flask --app api/app.py create-index
 ".elser_model_2" model not available, downloading it now
 Model downloaded, starting deployment
 Loading data from ./data/data.json
@@ -121,13 +126,30 @@ and retry.
 
 Now, run the app, which listens on http://localhost:4000
 ```bash
-$ dotenv run -- flask run
- * Serving Flask app 'api/app.py'
+$ dotenv run -- python api/app.py
+ * Serving Flask app 'app'
  * Debug mode: off
 ```
 
 ## Customizing the app
 
+### Updating package versions
+
+To update package versions, recreate [requirements.txt](requirements.txt) and
+reinstall like this. Once checked in, any commands above will use updates.
+
+```bash
+rm -rf .venv
+python3 -m venv .venv
+source .venv/bin/activate
+# Install dev requirements for pip-compile
+pip install pip-tools
+# Recreate requirements.txt
+pip-compile
+# Install main dependencies
+pip install -r requirements.txt
+```
+
 ### Indexing your own data
 
 The ingesting logic is stored in [data/index_data.py](data/index_data.py). This
diff --git a/example-apps/chatbot-rag-app/api/app.py b/example-apps/chatbot-rag-app/api/app.py
index 826b6349..a94f4a24 100644
--- a/example-apps/chatbot-rag-app/api/app.py
+++ b/example-apps/chatbot-rag-app/api/app.py
@@ -37,6 +37,5 @@ def create_index():
     index_data.main()
 
 
-# Unless we run through flask, we can miss critical settings or telemetry signals.
 if __name__ == "__main__":
-    raise RuntimeError("Run via the parent directory: 'flask run'")
+    app.run(host="0.0.0.0", port=4000, debug=False)
diff --git a/example-apps/chatbot-rag-app/docker-compose.yml b/example-apps/chatbot-rag-app/docker-compose.yml
index 634fa831..87996b95 100644
--- a/example-apps/chatbot-rag-app/docker-compose.yml
+++ b/example-apps/chatbot-rag-app/docker-compose.yml
@@ -4,9 +4,10 @@ services:
       context: .
     container_name: ingest-data
     restart: 'no'
+    environment:
+      FLASK_APP: api/app.py
     env_file:
       - .env
-      - .flaskenv
     command: flask create-index
 
   api-frontend:
@@ -18,6 +19,5 @@ services:
       context: .
     env_file:
       - .env
-      - .flaskenv
     ports:
       - "4000:4000"
diff --git a/example-apps/chatbot-rag-app/env.example b/example-apps/chatbot-rag-app/env.example
index d078293f..169c3c96 100644
--- a/example-apps/chatbot-rag-app/env.example
+++ b/example-apps/chatbot-rag-app/env.example
@@ -28,10 +28,10 @@ ES_INDEX_CHAT_HISTORY=workplace-app-docs-chat-history
 
 # Uncomment and complete if you want to use Bedrock LLM
 # LLM_TYPE=bedrock
-# AWS_ACCESS_KEY=
-# AWS_SECRET_KEY=
-# AWS_REGION=
-# AWS_MODEL_ID=
+# AWS_ACCESS_KEY_ID=
+# AWS_SECRET_ACCESS_KEY=
+# AWS_DEFAULT_REGION=
+# CHAT_MODEL=amazon.titan-text-lite-v1
 
 # Uncomment and complete if you want to use Vertex AI
 # LLM_TYPE=vertex
diff --git a/example-apps/chatbot-rag-app/llm_integrations.py b/example-apps/chatbot-rag-app/llm_integrations.py
new file mode 100644
index 00000000..1131b2e8
--- /dev/null
+++ b/example-apps/chatbot-rag-app/llm_integrations.py
@@ -0,0 +1,69 @@
+import os
+
+import vertexai
+from langchain_aws import ChatBedrock
+from langchain_cohere import ChatCohere
+from langchain_google_vertexai import ChatVertexAI
+from langchain_mistralai import ChatMistralAI
+from langchain_openai import AzureChatOpenAI, ChatOpenAI
+
+LLM_TYPE = os.getenv("LLM_TYPE", "openai")
+
+
+def init_openai_chat(temperature):
+    return ChatOpenAI(model=os.getenv("CHAT_MODEL"), streaming=True, temperature=temperature)
+
+
+def init_vertex_chat(temperature):
+    VERTEX_PROJECT_ID = os.getenv("VERTEX_PROJECT_ID")
+    VERTEX_REGION = os.getenv("VERTEX_REGION", "us-central1")
+    vertexai.init(project=VERTEX_PROJECT_ID, location=VERTEX_REGION)
+    return ChatVertexAI(streaming=True, temperature=temperature)
+
+
+def init_azure_chat(temperature):
+    return AzureChatOpenAI(model=os.getenv("CHAT_DEPLOYMENT"), streaming=True, temperature=temperature)
+
+
+def init_bedrock(temperature):
+    return ChatBedrock(model_id=os.getenv("CHAT_MODEL"), streaming=True, model_kwargs={"temperature": temperature})
+
+
+def init_mistral_chat(temperature):
+    MISTRAL_API_ENDPOINT = os.getenv("MISTRAL_API_ENDPOINT")
+    MISTRAL_API_KEY = os.getenv("MISTRAL_API_KEY")
+    MISTRAL_MODEL = os.getenv("MISTRAL_MODEL", "Mistral-large")
+    kwargs = {
+        "mistral_api_key": MISTRAL_API_KEY,
+        "temperature": temperature,
+    }
+    if MISTRAL_API_ENDPOINT:
+        kwargs["endpoint"] = MISTRAL_API_ENDPOINT
+    if MISTRAL_MODEL:
+        kwargs["model"] = MISTRAL_MODEL
+    return ChatMistralAI(**kwargs)
+
+
+def init_cohere_chat(temperature):
+    COHERE_API_KEY = os.getenv("COHERE_API_KEY")
+    COHERE_MODEL = os.getenv("COHERE_MODEL")
+    return ChatCohere(cohere_api_key=COHERE_API_KEY, model=COHERE_MODEL, temperature=temperature)
+
+
+MAP_LLM_TYPE_TO_CHAT_MODEL = {
+    "azure": init_azure_chat,
+    "bedrock": init_bedrock,
+    "openai": init_openai_chat,
+    "vertex": init_vertex_chat,
+    "mistral": init_mistral_chat,
+    "cohere": init_cohere_chat,
+}
+
+
+def get_llm(temperature=0):
+    if LLM_TYPE not in MAP_LLM_TYPE_TO_CHAT_MODEL:
+        raise Exception(
+            "LLM type not found. Please set LLM_TYPE to one of: " + ", ".join(MAP_LLM_TYPE_TO_CHAT_MODEL.keys()) + "."
+        )
+
+    return MAP_LLM_TYPE_TO_CHAT_MODEL[LLM_TYPE](temperature=temperature)
diff --git a/example-apps/chatbot-rag-app/requirements.txt b/example-apps/chatbot-rag-app/requirements.txt
index 21fea6a1..2e2f2f04 100644
--- a/example-apps/chatbot-rag-app/requirements.txt
+++ b/example-apps/chatbot-rag-app/requirements.txt
@@ -1,5 +1,5 @@
 #
-# This file is autogenerated by pip-compile with Python 3.12
+# This file is autogenerated by pip-compile with Python 3.13
 # by the following command:
 #
 #    pip-compile
@@ -352,7 +352,6 @@ types-requests==2.32.0.20241016
     # via cohere
 typing-extensions==4.12.2
     # via
-    #   anyio
     #   cohere
     #   huggingface-hub
     #   langchain-core