diff --git a/packages/create-llama/templates/.gitignore b/packages/create-llama/templates/.gitignore
new file mode 100644
index 0000000000000000000000000000000000000000..ec6c67b630467343abb46cfeea0535ce4b339554
--- /dev/null
+++ b/packages/create-llama/templates/.gitignore
@@ -0,0 +1,3 @@
+__pycache__
+poetry.lock
+storage
diff --git a/packages/create-llama/templates/streaming/fastapi/README-template.md b/packages/create-llama/templates/streaming/fastapi/README-template.md
new file mode 100644
index 0000000000000000000000000000000000000000..baa5fa63fcb1c07f8d74af3aa2eabd7bb493fda2
--- /dev/null
+++ b/packages/create-llama/templates/streaming/fastapi/README-template.md
@@ -0,0 +1,36 @@
+This is a [LlamaIndex](https://www.llamaindex.ai/) project using [FastAPI](https://fastapi.tiangolo.com/) bootstrapped with [`create-llama`](https://github.com/run-llama/LlamaIndexTS/tree/main/packages/create-llama).
+
+## Getting Started
+
+First, setup the environment:
+
+```
+poetry install
+poetry shell
+```
+
+Second, run the development server:
+
+```
+python main.py
+```
+
+Then call the API endpoint `/api/chat` to see the result:
+
+```
+curl --location 'localhost:8000/api/chat' \
+--header 'Content-Type: application/json' \
+--data '{ "messages": [{ "role": "user", "content": "Hello" }] }'
+```
+
+You can start editing the API by modifying `app/api/routers/chat.py`. The endpoint auto-updates as you save the file.
+
+Open [http://localhost:8000/docs](http://localhost:8000/docs) with your browser to see the Swagger UI of the API.
+
+## Learn More
+
+To learn more about LlamaIndex, take a look at the following resources:
+
+- [LlamaIndex Documentation](https://docs.llamaindex.ai) - learn about LlamaIndex.
+
+You can check out [the LlamaIndex GitHub repository](https://github.com/run-llama/llama_index) - your feedback and contributions are welcome!
diff --git a/packages/create-llama/templates/streaming/fastapi/app/__init__.py b/packages/create-llama/templates/streaming/fastapi/app/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/packages/create-llama/templates/streaming/fastapi/app/api/__init__.py b/packages/create-llama/templates/streaming/fastapi/app/api/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/packages/create-llama/templates/streaming/fastapi/app/api/routers/__init__.py b/packages/create-llama/templates/streaming/fastapi/app/api/routers/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/packages/create-llama/templates/streaming/fastapi/app/api/routers/chat.py b/packages/create-llama/templates/streaming/fastapi/app/api/routers/chat.py
new file mode 100644
index 0000000000000000000000000000000000000000..bc9b5ed651efff1a91f659b8130004b0b90be6a9
--- /dev/null
+++ b/packages/create-llama/templates/streaming/fastapi/app/api/routers/chat.py
@@ -0,0 +1,79 @@
+import logging
+import os
+from typing import List
+from fastapi import APIRouter, Depends, HTTPException, Request, status
+from llama_index import (
+ StorageContext,
+ load_index_from_storage,
+ SimpleDirectoryReader,
+ VectorStoreIndex,
+)
+from llama_index.llms.base import MessageRole
+from pydantic import BaseModel
+from sse_starlette.sse import EventSourceResponse
+
+STORAGE_DIR = "./storage" # directory to cache the generated index
+DATA_DIR = "./data" # directory containing the documents to index
+
+chat_router = r = APIRouter()
+
+
+class Message(BaseModel):
+ role: MessageRole
+ content: str
+
+
+class _ChatData(BaseModel):
+ messages: List[Message]
+
+
+def get_index():
+ logger = logging.getLogger("uvicorn")
+ # check if storage already exists
+ if not os.path.exists(STORAGE_DIR):
+ logger.info("Creating new index")
+ # load the documents and create the index
+ documents = SimpleDirectoryReader(DATA_DIR).load_data()
+ index = VectorStoreIndex.from_documents(documents)
+ # store it for later
+ index.storage_context.persist(STORAGE_DIR)
+ logger.info(f"Finished creating new index. Stored in {STORAGE_DIR}")
+ else:
+ # load the existing index
+ logger.info(f"Loading index from {STORAGE_DIR}...")
+ storage_context = StorageContext.from_defaults(persist_dir=STORAGE_DIR)
+ index = load_index_from_storage(storage_context)
+ logger.info(f"Finished loading index from {STORAGE_DIR}")
+ return index
+
+
+@r.post("/")
+async def chat(
+ request: Request, data: _ChatData, index: VectorStoreIndex = Depends(get_index)
+) -> Message:
+ # check preconditions
+ if len(data.messages) == 0:
+ raise HTTPException(
+ status_code=status.HTTP_400_BAD_REQUEST,
+ detail="No messages provided",
+ )
+ lastMessage = data.messages.pop()
+ if lastMessage.role != MessageRole.USER:
+ raise HTTPException(
+ status_code=status.HTTP_400_BAD_REQUEST,
+ detail="Last message must be from user",
+ )
+
+ # query chat engine
+ chat_engine = index.as_chat_engine()
+ response = chat_engine.stream_chat(lastMessage.content, data.messages)
+
+ # stream response
+ async def event_generator():
+ for token in response.response_gen:
+ # If client closes connection, stop sending events
+ if await request.is_disconnected():
+ break
+ yield token
+
+ return EventSourceResponse(event_generator())
diff --git a/packages/create-llama/templates/streaming/fastapi/data/brk-2022.pdf b/packages/create-llama/templates/streaming/fastapi/data/brk-2022.pdf
new file mode 100644
index 0000000000000000000000000000000000000000..876ea6a37f0fe785dffc74434e805b65a802b91f
Binary files /dev/null and b/packages/create-llama/templates/streaming/fastapi/data/brk-2022.pdf differ
diff --git a/packages/create-llama/templates/streaming/fastapi/gitignore b/packages/create-llama/templates/streaming/fastapi/gitignore
new file mode 100644
index 0000000000000000000000000000000000000000..069fcb4020566da83dbc398c06dce5bbe92aface
--- /dev/null
+++ b/packages/create-llama/templates/streaming/fastapi/gitignore
@@ -0,0 +1,2 @@
+__pycache__
+storage
diff --git a/packages/create-llama/templates/streaming/fastapi/main.py b/packages/create-llama/templates/streaming/fastapi/main.py
new file mode 100644
index 0000000000000000000000000000000000000000..e307354bc3b935a52d96c0e138a937969df4d4cf
--- /dev/null
+++ b/packages/create-llama/templates/streaming/fastapi/main.py
@@ -0,0 +1,23 @@
+import os
+import uvicorn
+from app.api.routers.chat import chat_router
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+
+app = FastAPI()
+
+origin = os.getenv("CORS_ORIGIN")
+if origin:
+ app.add_middleware(
+ CORSMiddleware,
+ allow_origins=[origin],
+ allow_credentials=True,
+ allow_methods=["*"],
+ allow_headers=["*"],
+ )
+
+app.include_router(chat_router, prefix="/api/chat")
+
+
+if __name__ == "__main__":
+ uvicorn.run(app="main:app", host="0.0.0.0", reload=True)
diff --git a/packages/create-llama/templates/streaming/fastapi/pyproject.toml b/packages/create-llama/templates/streaming/fastapi/pyproject.toml
new file mode 100644
index 0000000000000000000000000000000000000000..73d3cc51070a81f764999100cc7d7ec1df8f36c8
--- /dev/null
+++ b/packages/create-llama/templates/streaming/fastapi/pyproject.toml
@@ -0,0 +1,19 @@
+[tool.poetry]
+name = "llamaindex-fastapi-streaming"
+version = "0.1.0"
+description = ""
+authors = ["Marcus Schiesser <mail@marcusschiesser.de>"]
+readme = "README.md"
+
+[tool.poetry.dependencies]
+python = "^3.11,<3.12"
+fastapi = "^0.104.1"
+uvicorn = { extras = ["standard"], version = "^0.23.2" }
+llama-index = "^0.8.56"
+pypdf = "^3.17.0"
+sse-starlette = "^1.6.5"
+
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"
diff --git a/packages/create-llama/templates/streaming/fastapi/tests/__init__.py b/packages/create-llama/templates/streaming/fastapi/tests/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391