Release 0.1.7 (#93)

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

CHANGELOG.md

@@ -1,5 +1,40 @@
 # create-llama
 
+## 0.1.7
+
+### Patch Changes
+
+- 260d37a: Add system prompt env variable for TS
+- bbd5b8d: Fix postgres connection leaking issue
+- bb53425: Support HTTP proxies by setting the GLOBAL_AGENT_HTTP_PROXY env variable
+- 69c2e16: Fix streaming for Express
+- 7873bfb: Update Ollama provider to run with the base URL from the environment variable
+
+## 0.1.6
+
+### Patch Changes
+
+- 56537a1: Display PDF files in source nodes
+
+## 0.1.5
+
+### Patch Changes
+
+- 84db798: feat: support display latex in chat markdown
+
+## 0.1.4
+
+### Patch Changes
+
+- 0bc8e75: Use ingestion pipeline for dedicated vector stores (Python only)
+- cb1001d: Add ChromaDB vector store
+
+## 0.1.3
+
+### Patch Changes
+
+- 416073d: Directly import vector stores to work with NextJS
+
 ## 0.1.2
 
 ### Patch Changes

helpers/env-variables.ts

@@ -29,17 +29,20 @@ const renderEnvVar = (envVars: EnvVar[]): string => {
   );
 };
 
-const getVectorDBEnvs = (vectorDb?: TemplateVectorDB): EnvVar[] => {
-  if (!vectorDb) {
+const getVectorDBEnvs = (
+  vectorDb?: TemplateVectorDB,
+  framework?: TemplateFramework,
+): EnvVar[] => {
+  if (!vectorDb || !framework) {
     return [];
   }
   switch (vectorDb) {
     case "mongo":
       return [
         {
-          name: "MONGO_URI",
+          name: "MONGODB_URI",
           description:
-            "For generating a connection URI, see https://docs.timescale.com/use-timescale/latest/services/create-a-service\nThe MongoDB connection URI.",
+            "For generating a connection URI, see https://www.mongodb.com/docs/manual/reference/connection-string/ \nThe MongoDB connection URI.",
         },
         {
           name: "MONGODB_DATABASE",
@@ -129,6 +132,31 @@ const getVectorDBEnvs = (vectorDb?: TemplateVectorDB): EnvVar[] => {
             "Optional API key for authenticating requests to Qdrant.",
         },
       ];
+    case "chroma":
+      const envs = [
+        {
+          name: "CHROMA_COLLECTION",
+          description: "The name of the collection in your Chroma database",
+        },
+        {
+          name: "CHROMA_HOST",
+          description: "The API endpoint for your Chroma database",
+        },
+        {
+          name: "CHROMA_PORT",
+          description: "The port for your Chroma database",
+        },
+      ];
+      // TS Version doesn't support config local storage path
+      if (framework === "fastapi") {
+        envs.push({
+          name: "CHROMA_PATH",
+          description: `The local path to the Chroma database. 
+Specify this if you are using a local Chroma database. 
+Otherwise, use CHROMA_HOST and CHROMA_PORT config above`,
+        });
+      }
+      return envs;
     default:
       return [];
   }
@@ -191,6 +219,15 @@ const getModelEnvs = (modelConfig: ModelConfig): EnvVar[] => {
           },
         ]
       : []),
+    ...(modelConfig.provider === "ollama"
+      ? [
+          {
+            name: "OLLAMA_BASE_URL",
+            description:
+              "The base URL for the Ollama API. Eg: http://127.0.0.1:11434",
+          },
+        ]
+      : []),
   ];
 };
 
@@ -212,13 +249,6 @@ const getFrameworkEnvs = (
       description: "The port to start the backend app.",
       value: port?.toString() || "8000",
     },
-    // TODO: Once LlamaIndexTS supports string templates, move this to `getEngineEnvs`
-    {
-      name: "SYSTEM_PROMPT",
-      description: `Custom system prompt.
-Example:
-SYSTEM_PROMPT="You are a helpful assistant who helps users with their questions."`,
-    },
   ];
 };
 
@@ -230,6 +260,12 @@ const getEngineEnvs = (): EnvVar[] => {
         "The number of similar embeddings to return when retrieving documents.",
       value: "3",
     },
+    {
+      name: "SYSTEM_PROMPT",
+      description: `Custom system prompt.
+Example:
+SYSTEM_PROMPT="You are a helpful assistant who helps users with their questions."`,
+    },
   ];
 };
 
@@ -257,7 +293,7 @@ export const createBackendEnvFile = async (
     // Add engine environment variables
     ...getEngineEnvs(),
     // Add vector database environment variables
-    ...getVectorDBEnvs(opts.vectorDb),
+    ...getVectorDBEnvs(opts.vectorDb, opts.framework),
     ...getFrameworkEnvs(opts.framework, opts.port),
   ];
   // Render and write env file

helpers/proxy.ts

@@ -0,0 +1,8 @@
+/* Function to conditionally load the global-agent/bootstrap module */
+export async function initializeGlobalAgent() {
+  if (process.env.GLOBAL_AGENT_HTTP_PROXY) {
+    /* Dynamically import global-agent/bootstrap */
+    await import("global-agent/bootstrap");
+    console.log("Proxy enabled via global-agent.");
+  }
+}

helpers/python.ts

@@ -70,6 +70,20 @@ const getAdditionalDependencies = (
       });
       break;
     }
+    case "qdrant": {
+      dependencies.push({
+        name: "llama-index-vector-stores-qdrant",
+        version: "^0.2.8",
+      });
+      break;
+    }
+    case "chroma": {
+      dependencies.push({
+        name: "llama-index-vector-stores-chroma",
+        version: "^0.1.8",
+      });
+      break;
+    }
   }
 
   // Add data source dependencies

helpers/types.ts

@@ -20,7 +20,8 @@ export type TemplateVectorDB =
   | "pinecone"
   | "milvus"
   | "astra"
-  | "qdrant";
+  | "qdrant"
+  | "chroma";
 export type TemplatePostInstallAction =
   | "none"
   | "VSCode"

index.ts

@@ -12,12 +12,16 @@ import { createApp } from "./create-app";
 import { getDataSources } from "./helpers/datasources";
 import { getPkgManager } from "./helpers/get-pkg-manager";
 import { isFolderEmpty } from "./helpers/is-folder-empty";
+import { initializeGlobalAgent } from "./helpers/proxy";
 import { runApp } from "./helpers/run-app";
 import { getTools } from "./helpers/tools";
 import { validateNpmName } from "./helpers/validate-pkg";
 import packageJson from "./package.json";
 import { QuestionArgs, askQuestions, onPromptState } from "./questions";
 
+// Run the initialization function
+initializeGlobalAgent();
+
 let projectPath: string = "";
 
 const handleSigTerm = () => process.exit(0);

package.json

@@ -1,6 +1,6 @@
 {
   "name": "create-llama",
-  "version": "0.1.2",
+  "version": "0.1.7",
   "description": "Create LlamaIndex-powered apps with one command",
   "keywords": [
     "rag",
@@ -52,6 +52,7 @@
     "cross-spawn": "7.0.3",
     "fast-glob": "3.3.1",
     "fs-extra": "11.2.0",
+    "global-agent": "^3.0.0",
     "got": "10.7.0",
     "ollama": "^0.5.0",
     "ora": "^8.0.1",

questions.ts

@@ -97,6 +97,7 @@ const getVectorDbChoices = (framework: TemplateFramework) => {
     { title: "Milvus", value: "milvus" },
     { title: "Astra", value: "astra" },
     { title: "Qdrant", value: "qdrant" },
+    { title: "ChromaDB", value: "chroma" },
   ];
 
   const vectordbLang = framework === "fastapi" ? "python" : "typescript";

templates/components/engines/typescript/agent/chat.ts

@@ -41,5 +41,6 @@ export async function createChatEngine() {
 
   return new OpenAIAgent({
     tools,
+    systemPrompt: process.env.SYSTEM_PROMPT,
   });
 }

templates/components/engines/typescript/chat/chat.ts

@@ -16,5 +16,6 @@ export async function createChatEngine() {
   return new ContextChatEngine({
     chatModel: Settings.llm,
     retriever,
+    systemPrompt: process.env.SYSTEM_PROMPT,
   });
 }

templates/components/loaders/python/file.py

@@ -1,7 +1,10 @@
 import os
+import logging
 from llama_parse import LlamaParse
 from pydantic import BaseModel, validator
 
+logger = logging.getLogger(__name__)
+
 
 class FileLoaderConfig(BaseModel):
     data_dir: str = "data"
@@ -27,11 +30,28 @@ def llama_parse_parser():
 def get_file_documents(config: FileLoaderConfig):
     from llama_index.core.readers import SimpleDirectoryReader
 
-    reader = SimpleDirectoryReader(
-        config.data_dir,
-        recursive=True,
-    )
-    if config.use_llama_parse:
-        parser = llama_parse_parser()
-        reader.file_extractor = {".pdf": parser}
-    return reader.load_data()
+    try:
+        reader = SimpleDirectoryReader(
+            config.data_dir,
+            recursive=True,
+            filename_as_id=True,
+        )
+        if config.use_llama_parse:
+            parser = llama_parse_parser()
+            reader.file_extractor = {".pdf": parser}
+        return reader.load_data()
+    except ValueError as e:
+        import sys, traceback
+
+        # Catch the error if the data dir is empty
+        # and return as empty document list
+        _, _, exc_traceback = sys.exc_info()
+        function_name = traceback.extract_tb(exc_traceback)[-1].name
+        if function_name == "_add_files":
+            logger.warning(
+                f"Failed to load file documents, error message: {e} . Return as empty document list."
+            )
+            return []
+        else:
+            # Raise the error if it is not the case of empty data dir
+            raise e

templates/components/vectordbs/python/astra/generate.py

@@ -1,37 +0,0 @@
-from dotenv import load_dotenv
-
-load_dotenv()
-
-import os
-import logging
-from llama_index.core.storage import StorageContext
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.vector_stores.astra_db import AstraDBVectorStore
-from app.settings import init_settings
-from app.engine.loaders import get_documents
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger()
-
-
-def generate_datasource():
-    init_settings()
-    logger.info("Creating new index")
-    documents = get_documents()
-    store = AstraDBVectorStore(
-        token=os.environ["ASTRA_DB_APPLICATION_TOKEN"],
-        api_endpoint=os.environ["ASTRA_DB_ENDPOINT"],
-        collection_name=os.environ["ASTRA_DB_COLLECTION"],
-        embedding_dimension=int(os.environ["EMBEDDING_DIM"]),
-    )
-    storage_context = StorageContext.from_defaults(vector_store=store)
-    VectorStoreIndex.from_documents(
-        documents,
-        storage_context=storage_context,
-        show_progress=True,  # this will show you a progress bar as the embeddings are created
-    )
-    logger.info(f"Successfully created embeddings in the AstraDB")
-
-
-if __name__ == "__main__":
-    generate_datasource()

templates/components/vectordbs/python/astra/index.py

@@ -1,21 +0,0 @@
-import logging
-import os
-
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.vector_stores.astra_db import AstraDBVectorStore
-
-
-logger = logging.getLogger("uvicorn")
-
-
-def get_index():
-    logger.info("Connecting to index from AstraDB...")
-    store = AstraDBVectorStore(
-        token=os.environ["ASTRA_DB_APPLICATION_TOKEN"],
-        api_endpoint=os.environ["ASTRA_DB_ENDPOINT"],
-        collection_name=os.environ["ASTRA_DB_COLLECTION"],
-        embedding_dimension=int(os.environ["EMBEDDING_DIM"]),
-    )
-    index = VectorStoreIndex.from_vector_store(store)
-    logger.info("Finished connecting to index from AstraDB.")
-    return index

templates/components/vectordbs/python/astra/vectordb.py

@@ -0,0 +1,20 @@
+import os
+from llama_index.vector_stores.astra_db import AstraDBVectorStore
+
+
+def get_vector_store():
+    endpoint = os.getenv("ASTRA_DB_ENDPOINT")
+    token = os.getenv("ASTRA_DB_APPLICATION_TOKEN")
+    collection = os.getenv("ASTRA_DB_COLLECTION")
+    if not endpoint or not token or not collection:
+        raise ValueError(
+            "Please config ASTRA_DB_ENDPOINT, ASTRA_DB_APPLICATION_TOKEN and ASTRA_DB_COLLECTION"
+            " to your environment variables or config them in the .env file"
+        )
+    store = AstraDBVectorStore(
+        token=token,
+        api_endpoint=endpoint,
+        collection_name=collection,
+        embedding_dimension=int(os.getenv("EMBEDDING_DIM")),
+    )
+    return store

templates/components/vectordbs/python/chroma/vectordb.py

@@ -0,0 +1,24 @@
+import os
+from llama_index.vector_stores.chroma import ChromaVectorStore
+
+
+def get_vector_store():
+    collection_name = os.getenv("CHROMA_COLLECTION", "default")
+    chroma_path = os.getenv("CHROMA_PATH")
+    # if CHROMA_PATH is set, use a local ChromaVectorStore from the path
+    # otherwise, use a remote ChromaVectorStore (ChromaDB Cloud is not supported yet)
+    if chroma_path:
+        store = ChromaVectorStore.from_params(
+            persist_dir=chroma_path, collection_name=collection_name
+        )
+    else:
+        if not os.getenv("CHROMA_HOST") or not os.getenv("CHROMA_PORT"):
+            raise ValueError(
+                "Please provide either CHROMA_PATH or CHROMA_HOST and CHROMA_PORT"
+            )
+        store = ChromaVectorStore.from_params(
+            host=os.getenv("CHROMA_HOST"),
+            port=int(os.getenv("CHROMA_PORT")),
+            collection_name=collection_name,
+        )
+    return store

templates/components/vectordbs/python/milvus/generate.py

@@ -1,39 +0,0 @@
-from dotenv import load_dotenv
-
-load_dotenv()
-
-import os
-import logging
-from llama_index.core.storage import StorageContext
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.vector_stores.milvus import MilvusVectorStore
-from app.settings import init_settings
-from app.engine.loaders import get_documents
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger()
-
-
-def generate_datasource():
-    init_settings()
-    logger.info("Creating new index")
-    # load the documents and create the index
-    documents = get_documents()
-    store = MilvusVectorStore(
-        uri=os.environ["MILVUS_ADDRESS"],
-        user=os.getenv("MILVUS_USERNAME"),
-        password=os.getenv("MILVUS_PASSWORD"),
-        collection_name=os.getenv("MILVUS_COLLECTION"),
-        dim=int(os.getenv("EMBEDDING_DIM")),
-    )
-    storage_context = StorageContext.from_defaults(vector_store=store)
-    VectorStoreIndex.from_documents(
-        documents,
-        storage_context=storage_context,
-        show_progress=True,  # this will show you a progress bar as the embeddings are created
-    )
-    logger.info(f"Successfully created embeddings in the Milvus")
-
-
-if __name__ == "__main__":
-    generate_datasource()

templates/components/vectordbs/python/milvus/index.py

@@ -1,22 +0,0 @@
-import logging
-import os
-
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.vector_stores.milvus import MilvusVectorStore
-
-
-logger = logging.getLogger("uvicorn")
-
-
-def get_index():
-    logger.info("Connecting to index from Milvus...")
-    store = MilvusVectorStore(
-        uri=os.getenv("MILVUS_ADDRESS"),
-        user=os.getenv("MILVUS_USERNAME"),
-        password=os.getenv("MILVUS_PASSWORD"),
-        collection_name=os.getenv("MILVUS_COLLECTION"),
-        dim=int(os.getenv("EMBEDDING_DIM")),
-    )
-    index = VectorStoreIndex.from_vector_store(store)
-    logger.info("Finished connecting to index from Milvus.")
-    return index

templates/components/vectordbs/python/milvus/vectordb.py

@@ -0,0 +1,20 @@
+import os
+from llama_index.vector_stores.milvus import MilvusVectorStore
+
+
+def get_vector_store():
+    address = os.getenv("MILVUS_ADDRESS")
+    collection = os.getenv("MILVUS_COLLECTION")
+    if not address or not collection:
+        raise ValueError(
+            "Please set MILVUS_ADDRESS and MILVUS_COLLECTION to your environment variables"
+            " or config them in the .env file"
+        )
+    store = MilvusVectorStore(
+        uri=address,
+        user=os.getenv("MILVUS_USERNAME"),
+        password=os.getenv("MILVUS_PASSWORD"),
+        collection_name=collection,
+        dim=int(os.getenv("EMBEDDING_DIM")),
+    )
+    return store

templates/components/vectordbs/python/mongo/generate.py

@@ -1,43 +0,0 @@
-from dotenv import load_dotenv
-
-load_dotenv()
-
-import os
-import logging
-from llama_index.core.storage import StorageContext
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.vector_stores.mongodb import MongoDBAtlasVectorSearch
-from app.settings import init_settings
-from app.engine.loaders import get_documents
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger()
-
-
-def generate_datasource():
-    init_settings()
-    logger.info("Creating new index")
-    # load the documents and create the index
-    documents = get_documents()
-    store = MongoDBAtlasVectorSearch(
-        db_name=os.environ["MONGODB_DATABASE"],
-        collection_name=os.environ["MONGODB_VECTORS"],
-        index_name=os.environ["MONGODB_VECTOR_INDEX"],
-    )
-    storage_context = StorageContext.from_defaults(vector_store=store)
-    VectorStoreIndex.from_documents(
-        documents,
-        storage_context=storage_context,
-        show_progress=True,  # this will show you a progress bar as the embeddings are created
-    )
-    logger.info(
-        f"Successfully created embeddings in the MongoDB collection {os.environ['MONGODB_VECTORS']}"
-    )
-    logger.info(
-        """IMPORTANT: You can't query your index yet because you need to create a vector search index in MongoDB's UI now.
-See https://github.com/run-llama/mongodb-demo/tree/main?tab=readme-ov-file#create-a-vector-search-index"""
-    )
-
-
-if __name__ == "__main__":
-    generate_datasource()

templates/components/vectordbs/python/mongo/index.py

@@ -1,20 +0,0 @@
-import logging
-import os
-
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.vector_stores.mongodb import MongoDBAtlasVectorSearch
-
-
-logger = logging.getLogger("uvicorn")
-
-
-def get_index():
-    logger.info("Connecting to index from MongoDB...")
-    store = MongoDBAtlasVectorSearch(
-        db_name=os.environ["MONGODB_DATABASE"],
-        collection_name=os.environ["MONGODB_VECTORS"],
-        index_name=os.environ["MONGODB_VECTOR_INDEX"],
-    )
-    index = VectorStoreIndex.from_vector_store(store)
-    logger.info("Finished connecting to index from MongoDB.")
-    return index

templates/components/vectordbs/python/mongo/vectordb.py

@@ -0,0 +1,20 @@
+import os
+from llama_index.vector_stores.mongodb import MongoDBAtlasVectorSearch
+
+
+def get_vector_store():
+    db_uri = os.getenv("MONGODB_URI")
+    db_name = os.getenv("MONGODB_DATABASE")
+    collection_name = os.getenv("MONGODB_VECTORS")
+    index_name = os.getenv("MONGODB_VECTOR_INDEX")
+    if not db_uri or not db_name or not collection_name or not index_name:
+        raise ValueError(
+            "Please set MONGODB_URI, MONGODB_DATABASE, MONGODB_VECTORS, and MONGODB_VECTOR_INDEX"
+            " to your environment variables or config them in .env file"
+        )
+    store = MongoDBAtlasVectorSearch(
+        db_name=db_name,
+        collection_name=collection_name,
+        index_name=index_name,
+    )
+    return store

templates/components/vectordbs/python/none/constants.py

@@ -1 +0,0 @@
-STORAGE_DIR = "storage"  # directory to cache the generated index

templates/components/vectordbs/python/none/generate.py

@@ -2,11 +2,11 @@ from dotenv import load_dotenv
 
 load_dotenv()
 
+import os
 import logging
 from llama_index.core.indices import (
     VectorStoreIndex,
 )
-from app.engine.constants import STORAGE_DIR
 from app.engine.loaders import get_documents
 from app.settings import init_settings
 
@@ -18,14 +18,15 @@ logger = logging.getLogger()
 def generate_datasource():
     init_settings()
     logger.info("Creating new index")
+    storage_dir = os.environ.get("STORAGE_DIR", "storage")
     # load the documents and create the index
     documents = get_documents()
     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}")
+    index.storage_context.persist(storage_dir)
+    logger.info(f"Finished creating new index. Stored in {storage_dir}")
 
 
 if __name__ == "__main__":

templates/components/vectordbs/python/none/index.py

@@ -1,20 +1,30 @@
-import logging
 import os
+import logging
+from datetime import timedelta
 
-from app.engine.constants import STORAGE_DIR
+from cachetools import cached, TTLCache
 from llama_index.core.storage import StorageContext
 from llama_index.core.indices import load_index_from_storage
 
 logger = logging.getLogger("uvicorn")
 
 
+@cached(
+    TTLCache(maxsize=10, ttl=timedelta(minutes=5).total_seconds()),
+    key=lambda *args, **kwargs: "global_storage_context",
+)
+def get_storage_context(persist_dir: str) -> StorageContext:
+    return StorageContext.from_defaults(persist_dir=persist_dir)
+
+
 def get_index():
+    storage_dir = os.getenv("STORAGE_DIR", "storage")
     # check if storage already exists
-    if not os.path.exists(STORAGE_DIR):
+    if not os.path.exists(storage_dir):
         return None
     # load the existing index
-    logger.info(f"Loading index from {STORAGE_DIR}...")
-    storage_context = StorageContext.from_defaults(persist_dir=STORAGE_DIR)
+    logger.info(f"Loading index from {storage_dir}...")
+    storage_context = get_storage_context(storage_dir)
     index = load_index_from_storage(storage_context)
-    logger.info(f"Finished loading index from {STORAGE_DIR}")
+    logger.info(f"Finished loading index from {storage_dir}")
     return index

templates/components/vectordbs/python/pg/constants.py

@@ -1,2 +0,0 @@
-PGVECTOR_SCHEMA = "public"
-PGVECTOR_TABLE = "llamaindex_embedding"

templates/components/vectordbs/python/pg/generate.py

@@ -1,35 +0,0 @@
-from dotenv import load_dotenv
-
-load_dotenv()
-
-import logging
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.core.storage import StorageContext
-
-from app.engine.loaders import get_documents
-from app.settings import init_settings
-from app.engine.utils import init_pg_vector_store_from_env
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger()
-
-
-def generate_datasource():
-    init_settings()
-    logger.info("Creating new index")
-    # load the documents and create the index
-    documents = get_documents()
-    store = init_pg_vector_store_from_env()
-    storage_context = StorageContext.from_defaults(vector_store=store)
-    VectorStoreIndex.from_documents(
-        documents,
-        storage_context=storage_context,
-        show_progress=True,  # this will show you a progress bar as the embeddings are created
-    )
-    logger.info(
-        f"Successfully created embeddings in the PG vector store, schema={store.schema_name} table={store.table_name}"
-    )
-
-
-if __name__ == "__main__":
-    generate_datasource()

templates/components/vectordbs/python/pg/index.py

@@ -1,13 +0,0 @@
-import logging
-from llama_index.core.indices.vector_store import VectorStoreIndex
-from app.engine.utils import init_pg_vector_store_from_env
-
-logger = logging.getLogger("uvicorn")
-
-
-def get_index():
-    logger.info("Connecting to index from PGVector...")
-    store = init_pg_vector_store_from_env()
-    index = VectorStoreIndex.from_vector_store(store)
-    logger.info("Finished connecting to index from PGVector.")
-    return index

templates/components/vectordbs/python/pg/utils.py

@@ -1,27 +0,0 @@
-import os
-from llama_index.vector_stores.postgres import PGVectorStore
-from urllib.parse import urlparse
-from app.engine.constants import PGVECTOR_SCHEMA, PGVECTOR_TABLE
-
-
-def init_pg_vector_store_from_env():
-    original_conn_string = os.environ.get("PG_CONNECTION_STRING")
-    if original_conn_string is None or original_conn_string == "":
-        raise ValueError("PG_CONNECTION_STRING environment variable is not set.")
-
-    # The PGVectorStore requires both two connection strings, one for psycopg2 and one for asyncpg
-    # Update the configured scheme with the psycopg2 and asyncpg schemes
-    original_scheme = urlparse(original_conn_string).scheme + "://"
-    conn_string = original_conn_string.replace(
-        original_scheme, "postgresql+psycopg2://"
-    )
-    async_conn_string = original_conn_string.replace(
-        original_scheme, "postgresql+asyncpg://"
-    )
-
-    return PGVectorStore(
-        connection_string=conn_string,
-        async_connection_string=async_conn_string,
-        schema_name=PGVECTOR_SCHEMA,
-        table_name=PGVECTOR_TABLE,
-    )

templates/components/vectordbs/python/pg/vectordb.py

@@ -0,0 +1,37 @@
+import os
+from llama_index.vector_stores.postgres import PGVectorStore
+from urllib.parse import urlparse
+
+PGVECTOR_SCHEMA = "public"
+PGVECTOR_TABLE = "llamaindex_embedding"
+
+vector_store: PGVectorStore = None
+
+
+def get_vector_store():
+    global vector_store
+
+    if vector_store is None:
+        original_conn_string = os.environ.get("PG_CONNECTION_STRING")
+        if original_conn_string is None or original_conn_string == "":
+            raise ValueError("PG_CONNECTION_STRING environment variable is not set.")
+
+        # The PGVectorStore requires both two connection strings, one for psycopg2 and one for asyncpg
+        # Update the configured scheme with the psycopg2 and asyncpg schemes
+        original_scheme = urlparse(original_conn_string).scheme + "://"
+        conn_string = original_conn_string.replace(
+            original_scheme, "postgresql+psycopg2://"
+        )
+        async_conn_string = original_conn_string.replace(
+            original_scheme, "postgresql+asyncpg://"
+        )
+
+        vector_store = PGVectorStore(
+            connection_string=conn_string,
+            async_connection_string=async_conn_string,
+            schema_name=PGVECTOR_SCHEMA,
+            table_name=PGVECTOR_TABLE,
+            embed_dim=int(os.environ.get("EMBEDDING_DIM", 1024)),
+        )
+
+    return vector_store

templates/components/vectordbs/python/pinecone/generate.py

@@ -1,39 +0,0 @@
-from dotenv import load_dotenv
-
-load_dotenv()
-
-import os
-import logging
-from llama_index.core.storage import StorageContext
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.vector_stores.pinecone import PineconeVectorStore
-from app.settings import init_settings
-from app.engine.loaders import get_documents
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger()
-
-
-def generate_datasource():
-    init_settings()
-    logger.info("Creating new index")
-    # load the documents and create the index
-    documents = get_documents()
-    store = PineconeVectorStore(
-        api_key=os.environ["PINECONE_API_KEY"],
-        index_name=os.environ["PINECONE_INDEX_NAME"],
-        environment=os.environ["PINECONE_ENVIRONMENT"],
-    )
-    storage_context = StorageContext.from_defaults(vector_store=store)
-    VectorStoreIndex.from_documents(
-        documents,
-        storage_context=storage_context,
-        show_progress=True,  # this will show you a progress bar as the embeddings are created
-    )
-    logger.info(
-        f"Successfully created embeddings and save to your Pinecone index {os.environ['PINECONE_INDEX_NAME']}"
-    )
-
-
-if __name__ == "__main__":
-    generate_datasource()

templates/components/vectordbs/python/pinecone/index.py

@@ -1,20 +0,0 @@
-import logging
-import os
-
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.vector_stores.pinecone import PineconeVectorStore
-
-
-logger = logging.getLogger("uvicorn")
-
-
-def get_index():
-    logger.info("Connecting to index from Pinecone...")
-    store = PineconeVectorStore(
-        api_key=os.environ["PINECONE_API_KEY"],
-        index_name=os.environ["PINECONE_INDEX_NAME"],
-        environment=os.environ["PINECONE_ENVIRONMENT"],
-    )
-    index = VectorStoreIndex.from_vector_store(store)
-    logger.info("Finished connecting to index from Pinecone.")
-    return index

templates/components/vectordbs/python/pinecone/vectordb.py

@@ -0,0 +1,19 @@
+import os
+from llama_index.vector_stores.pinecone import PineconeVectorStore
+
+
+def get_vector_store():
+    api_key = os.getenv("PINECONE_API_KEY")
+    index_name = os.getenv("PINECONE_INDEX_NAME")
+    environment = os.getenv("PINECONE_ENVIRONMENT")
+    if not api_key or not index_name or not environment:
+        raise ValueError(
+            "Please set PINECONE_API_KEY, PINECONE_INDEX_NAME, and PINECONE_ENVIRONMENT"
+            " to your environment variables or config them in the .env file"
+        )
+    store = PineconeVectorStore(
+        api_key=api_key,
+        index_name=index_name,
+        environment=environment,
+    )
+    return store

templates/components/vectordbs/python/qdrant/generate.py

@@ -1,37 +0,0 @@
-import logging
-import os
-from app.engine.loaders import get_documents
-from app.settings import init_settings
-from dotenv import load_dotenv
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.core.storage import StorageContext
-from llama_index.vector_stores.qdrant import QdrantVectorStore
-load_dotenv()
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger()
-
-
-def generate_datasource():
-    init_settings()
-    logger.info("Creating new index with Qdrant")
-    # load the documents and create the index
-    documents = get_documents()
-    store = QdrantVectorStore(
-        collection_name=os.getenv("QDRANT_COLLECTION"),
-        url=os.getenv("QDRANT_URL"),
-        api_key=os.getenv("QDRANT_API_KEY"),
-    )
-    storage_context = StorageContext.from_defaults(vector_store=store)
-    VectorStoreIndex.from_documents(
-        documents,
-        storage_context=storage_context,
-        show_progress=True,  # this will show you a progress bar as the embeddings are created
-    )
-    logger.info(
-        f"Successfully uploaded documents to the {os.getenv('QDRANT_COLLECTION')} collection."
-    )
-
-
-if __name__ == "__main__":
-    generate_datasource()

templates/components/vectordbs/python/qdrant/index.py

@@ -1,20 +0,0 @@
-import logging
-import os
-
-from llama_index.core.indices import VectorStoreIndex
-from llama_index.vector_stores.qdrant import QdrantVectorStore
-
-
-logger = logging.getLogger("uvicorn")
-
-
-def get_index():
-    logger.info("Connecting to Qdrant collection..")
-    store = QdrantVectorStore(
-        collection_name=os.getenv("QDRANT_COLLECTION"),
-        url=os.getenv("QDRANT_URL"),
-        api_key=os.getenv("QDRANT_API_KEY"),
-    )
-    index = VectorStoreIndex.from_vector_store(store)
-    logger.info("Finished connecting to Qdrant collection.")
-    return index

templates/components/vectordbs/python/qdrant/vectordb.py

@@ -0,0 +1,19 @@
+import os
+from llama_index.vector_stores.qdrant import QdrantVectorStore
+
+
+def get_vector_store():
+    collection_name = os.getenv("QDRANT_COLLECTION")
+    url = os.getenv("QDRANT_URL")
+    api_key = os.getenv("QDRANT_API_KEY")
+    if not collection_name or not url:
+        raise ValueError(
+            "Please set QDRANT_COLLECTION, QDRANT_URL"
+            " to your environment variables or config them in the .env file"
+        )
+    store = QdrantVectorStore(
+        collection_name=collection_name,
+        url=url,
+        api_key=api_key,
+    )
+    return store

templates/components/vectordbs/typescript/astra/generate.ts

@@ -1,10 +1,7 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
 import * as dotenv from "dotenv";
-import {
-  AstraDBVectorStore,
-  VectorStoreIndex,
-  storageContextFromDefaults,
-} from "llamaindex";
+import { VectorStoreIndex, storageContextFromDefaults } from "llamaindex";
+import { AstraDBVectorStore } from "llamaindex/storage/vectorStore/AstraDBVectorStore";
 import { getDocuments } from "./loader";
 import { initSettings } from "./settings";
 import { checkRequiredEnvVars } from "./shared";

templates/components/vectordbs/typescript/astra/index.ts

@@ -1,5 +1,6 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
-import { AstraDBVectorStore, VectorStoreIndex } from "llamaindex";
+import { VectorStoreIndex } from "llamaindex";
+import { AstraDBVectorStore } from "llamaindex/storage/vectorStore/AstraDBVectorStore";
 import { checkRequiredEnvVars } from "./shared";
 
 export async function getDataSource() {

templates/components/vectordbs/typescript/chroma/generate.ts

@@ -0,0 +1,37 @@
+/* eslint-disable turbo/no-undeclared-env-vars */
+import * as dotenv from "dotenv";
+import { VectorStoreIndex, storageContextFromDefaults } from "llamaindex";
+import { ChromaVectorStore } from "llamaindex/storage/vectorStore/ChromaVectorStore";
+import { getDocuments } from "./loader";
+import { initSettings } from "./settings";
+import { checkRequiredEnvVars } from "./shared";
+
+dotenv.config();
+
+async function loadAndIndex() {
+  // load objects from storage and convert them into LlamaIndex Document objects
+  const documents = await getDocuments();
+
+  // create vector store
+  const chromaUri = `http://${process.env.CHROMA_HOST}:${process.env.CHROMA_PORT}`;
+
+  const vectorStore = new ChromaVectorStore({
+    collectionName: process.env.CHROMA_COLLECTION,
+    chromaClientParams: { path: chromaUri },
+  });
+
+  // create index from all the Documentss and store them in Pinecone
+  console.log("Start creating embeddings...");
+  const storageContext = await storageContextFromDefaults({ vectorStore });
+  await VectorStoreIndex.fromDocuments(documents, { storageContext });
+  console.log(
+    "Successfully created embeddings and save to your ChromaDB index.",
+  );
+}
+
+(async () => {
+  checkRequiredEnvVars();
+  initSettings();
+  await loadAndIndex();
+  console.log("Finished generating storage.");
+})();

templates/components/vectordbs/typescript/chroma/index.ts

@@ -0,0 +1,16 @@
+/* eslint-disable turbo/no-undeclared-env-vars */
+import { VectorStoreIndex } from "llamaindex";
+import { ChromaVectorStore } from "llamaindex/storage/vectorStore/ChromaVectorStore";
+import { checkRequiredEnvVars } from "./shared";
+
+export async function getDataSource() {
+  checkRequiredEnvVars();
+  const chromaUri = `http://${process.env.CHROMA_HOST}:${process.env.CHROMA_PORT}`;
+
+  const store = new ChromaVectorStore({
+    collectionName: process.env.CHROMA_COLLECTION,
+    chromaClientParams: { path: chromaUri },
+  });
+
+  return await VectorStoreIndex.fromVectorStore(store);
+}

templates/components/vectordbs/typescript/chroma/shared.ts

@@ -0,0 +1,18 @@
+const REQUIRED_ENV_VARS = ["CHROMA_COLLECTION", "CHROMA_HOST", "CHROMA_PORT"];
+
+export function checkRequiredEnvVars() {
+  const missingEnvVars = REQUIRED_ENV_VARS.filter((envVar) => {
+    return !process.env[envVar];
+  });
+
+  if (missingEnvVars.length > 0) {
+    console.log(
+      `The following environment variables are required but missing: ${missingEnvVars.join(
+        ", ",
+      )}`,
+    );
+    throw new Error(
+      `Missing environment variables: ${missingEnvVars.join(", ")}`,
+    );
+  }
+}

templates/components/vectordbs/typescript/milvus/generate.ts

@@ -1,10 +1,7 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
 import * as dotenv from "dotenv";
-import {
-  MilvusVectorStore,
-  VectorStoreIndex,
-  storageContextFromDefaults,
-} from "llamaindex";
+import { VectorStoreIndex, storageContextFromDefaults } from "llamaindex";
+import { MilvusVectorStore } from "llamaindex/storage/vectorStore/MilvusVectorStore";
 import { getDocuments } from "./loader";
 import { initSettings } from "./settings";
 import { checkRequiredEnvVars, getMilvusClient } from "./shared";

templates/components/vectordbs/typescript/milvus/index.ts

@@ -1,4 +1,5 @@
-import { MilvusVectorStore, VectorStoreIndex } from "llamaindex";
+import { VectorStoreIndex } from "llamaindex";
+import { MilvusVectorStore } from "llamaindex/storage/vectorStore/MilvusVectorStore";
 import { checkRequiredEnvVars, getMilvusClient } from "./shared";
 
 export async function getDataSource() {

templates/components/vectordbs/typescript/mongo/generate.ts

@@ -1,10 +1,7 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
 import * as dotenv from "dotenv";
-import {
-  MongoDBAtlasVectorSearch,
-  VectorStoreIndex,
-  storageContextFromDefaults,
-} from "llamaindex";
+import { VectorStoreIndex, storageContextFromDefaults } from "llamaindex";
+import { MongoDBAtlasVectorSearch } from "llamaindex/storage/vectorStore/MongoDBAtlasVectorSearch";
 import { MongoClient } from "mongodb";
 import { getDocuments } from "./loader";
 import { initSettings } from "./settings";
@@ -12,7 +9,7 @@ import { checkRequiredEnvVars } from "./shared";
 
 dotenv.config();
 
-const mongoUri = process.env.MONGO_URI!;
+const mongoUri = process.env.MONGODB_URI!;
 const databaseName = process.env.MONGODB_DATABASE!;
 const vectorCollectionName = process.env.MONGODB_VECTORS!;
 const indexName = process.env.MONGODB_VECTOR_INDEX;

templates/components/vectordbs/typescript/mongo/index.ts

@@ -1,5 +1,6 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
-import { MongoDBAtlasVectorSearch, VectorStoreIndex } from "llamaindex";
+import { VectorStoreIndex } from "llamaindex";
+import { MongoDBAtlasVectorSearch } from "llamaindex/storage/vectorStore/MongoDBAtlasVectorSearch";
 import { MongoClient } from "mongodb";
 import { checkRequiredEnvVars } from "./shared";
 

templates/components/vectordbs/typescript/mongo/shared.ts

@@ -1,5 +1,5 @@
 const REQUIRED_ENV_VARS = [
-  "MONGO_URI",
+  "MONGODB_URI",
   "MONGODB_DATABASE",
   "MONGODB_VECTORS",
   "MONGODB_VECTOR_INDEX",

templates/components/vectordbs/typescript/pg/generate.ts

@@ -1,10 +1,7 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
 import * as dotenv from "dotenv";
-import {
-  PGVectorStore,
-  VectorStoreIndex,
-  storageContextFromDefaults,
-} from "llamaindex";
+import { VectorStoreIndex, storageContextFromDefaults } from "llamaindex";
+import { PGVectorStore } from "llamaindex/storage/vectorStore/PGVectorStore";
 import { getDocuments } from "./loader";
 import { initSettings } from "./settings";
 import {

templates/components/vectordbs/typescript/pg/index.ts

@@ -1,5 +1,6 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
-import { PGVectorStore, VectorStoreIndex } from "llamaindex";
+import { VectorStoreIndex } from "llamaindex";
+import { PGVectorStore } from "llamaindex/storage/vectorStore/PGVectorStore";
 import {
   PGVECTOR_SCHEMA,
   PGVECTOR_TABLE,

templates/components/vectordbs/typescript/pinecone/generate.ts

@@ -1,10 +1,7 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
 import * as dotenv from "dotenv";
-import {
-  PineconeVectorStore,
-  VectorStoreIndex,
-  storageContextFromDefaults,
-} from "llamaindex";
+import { VectorStoreIndex, storageContextFromDefaults } from "llamaindex";
+import { PineconeVectorStore } from "llamaindex/storage/vectorStore/PineconeVectorStore";
 import { getDocuments } from "./loader";
 import { initSettings } from "./settings";
 import { checkRequiredEnvVars } from "./shared";

templates/components/vectordbs/typescript/pinecone/index.ts

@@ -1,5 +1,6 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
-import { PineconeVectorStore, VectorStoreIndex } from "llamaindex";
+import { VectorStoreIndex } from "llamaindex";
+import { PineconeVectorStore } from "llamaindex/storage/vectorStore/PineconeVectorStore";
 import { checkRequiredEnvVars } from "./shared";
 
 export async function getDataSource() {

templates/components/vectordbs/typescript/qdrant/generate.ts

@@ -1,10 +1,7 @@
 /* eslint-disable turbo/no-undeclared-env-vars */
 import * as dotenv from "dotenv";
-import {
-  QdrantVectorStore,
-  VectorStoreIndex,
-  storageContextFromDefaults,
-} from "llamaindex";
+import { VectorStoreIndex, storageContextFromDefaults } from "llamaindex";
+import { QdrantVectorStore } from "llamaindex/storage/vectorStore/QdrantVectorStore";
 import { getDocuments } from "./loader";
 import { initSettings } from "./settings";
 import { checkRequiredEnvVars, getQdrantClient } from "./shared";
@@ -18,7 +15,10 @@ async function loadAndIndex() {
   const documents = await getDocuments();
 
   // Connect to Qdrant
-  const vectorStore = new QdrantVectorStore(collectionName, getQdrantClient());
+  const vectorStore = new QdrantVectorStore({
+    collectionName,
+    client: getQdrantClient(),
+  });
 
   const storageContext = await storageContextFromDefaults({ vectorStore });
   await VectorStoreIndex.fromDocuments(documents, {

templates/components/vectordbs/typescript/qdrant/index.ts

@@ -1,5 +1,6 @@
 import * as dotenv from "dotenv";
-import { QdrantVectorStore, VectorStoreIndex } from "llamaindex";
+import { VectorStoreIndex } from "llamaindex";
+import { QdrantVectorStore } from "llamaindex/storage/vectorStore/QdrantVectorStore";
 import { checkRequiredEnvVars, getQdrantClient } from "./shared";
 
 dotenv.config();
@@ -7,7 +8,10 @@ dotenv.config();
 export async function getDataSource() {
   checkRequiredEnvVars();
   const collectionName = process.env.QDRANT_COLLECTION;
-  const store = new QdrantVectorStore(collectionName, getQdrantClient());
+  const store = new QdrantVectorStore({
+    collectionName,
+    client: getQdrantClient(),
+  });
 
   return await VectorStoreIndex.fromVectorStore(store);
 }

templates/types/streaming/express/index.ts

@@ -31,6 +31,7 @@ if (isDevelopment) {
   console.warn("Production CORS origin not set, defaulting to no CORS.");
 }
 
+app.use("/api/data", express.static("data"));
 app.use(express.text());
 
 app.get("/", (req: Request, res: Response) => {

templates/types/streaming/express/package.json

@@ -1,20 +1,20 @@
 {
   "name": "llama-index-express-streaming",
   "version": "1.0.0",
-  "main": "dist/index.mjs",
+  "main": "dist/index.js",
   "scripts": {
     "format": "prettier --ignore-unknown --cache --check .",
     "format:write": "prettier --ignore-unknown --write .",
-    "build": "tsup index.ts --format esm --dts",
-    "start": "node dist/index.mjs",
-    "dev": "concurrently \"tsup index.ts --format esm --dts --watch\" \"nodemon -q dist/index.mjs\""
+    "build": "tsup index.ts --format cjs --dts",
+    "start": "node dist/index.js",
+    "dev": "concurrently \"tsup index.ts --format cjs --dts --watch\" \"nodemon -q dist/index.js\""
   },
   "dependencies": {
     "ai": "^3.0.21",
     "cors": "^2.8.5",
     "dotenv": "^16.3.1",
     "express": "^4.18.2",
-    "llamaindex": "0.3.7",
+    "llamaindex": "0.3.13",
     "pdf2json": "3.0.5",
     "ajv": "^8.12.0"
   },

templates/types/streaming/express/src/controllers/chat.controller.ts

@@ -64,9 +64,8 @@ export const chat = async (req: Request, res: Response) => {
         image_url: data?.imageUrl,
       },
     });
-    const processedStream = stream.pipeThrough(vercelStreamData.stream);
 
-    return streamToResponse(processedStream, res);
+    return streamToResponse(stream, res, {}, vercelStreamData);
   } catch (error) {
     console.error("[LlamaIndex]", error);
     return res.status(500).json({

templates/types/streaming/express/src/controllers/engine/settings.ts

@@ -56,11 +56,17 @@ function initOpenAI() {
 }
 
 function initOllama() {
+  const config = {
+    host: process.env.OLLAMA_BASE_URL ?? "http://127.0.0.1:11434",
+  };
+
   Settings.llm = new Ollama({
     model: process.env.MODEL ?? "",
+    config,
   });
   Settings.embedModel = new OllamaEmbedding({
     model: process.env.EMBEDDING_MODEL ?? "",
+    config,
   });
 }
 

templates/types/streaming/fastapi/app/engine/generate.py

@@ -0,0 +1,80 @@
+from dotenv import load_dotenv
+
+load_dotenv()
+
+import os
+import logging
+from llama_index.core.settings import Settings
+from llama_index.core.ingestion import IngestionPipeline
+from llama_index.core.node_parser import SentenceSplitter
+from llama_index.core.storage.docstore import SimpleDocumentStore
+from llama_index.core.storage import StorageContext
+from app.settings import init_settings
+from app.engine.loaders import get_documents
+from app.engine.vectordb import get_vector_store
+
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger()
+
+STORAGE_DIR = os.getenv("STORAGE_DIR", "storage")
+
+
+def get_doc_store():
+
+    # If the storage directory is there, load the document store from it.
+    # If not, set up an in-memory document store since we can't load from a directory that doesn't exist.
+    if os.path.exists(STORAGE_DIR):
+        return SimpleDocumentStore.from_persist_dir(STORAGE_DIR)
+    else:
+        return SimpleDocumentStore()
+
+
+def run_pipeline(docstore, vector_store, documents):
+    pipeline = IngestionPipeline(
+        transformations=[
+            SentenceSplitter(
+                chunk_size=Settings.chunk_size,
+                chunk_overlap=Settings.chunk_overlap,
+            ),
+            Settings.embed_model,
+        ],
+        docstore=docstore,
+        docstore_strategy="upserts_and_delete",
+        vector_store=vector_store,
+    )
+
+    # Run the ingestion pipeline and store the results
+    nodes = pipeline.run(show_progress=True, documents=documents)
+
+    return nodes
+
+
+def persist_storage(docstore, vector_store):
+    storage_context = StorageContext.from_defaults(
+        docstore=docstore,
+        vector_store=vector_store,
+    )
+    storage_context.persist(STORAGE_DIR)
+
+
+def generate_datasource():
+    init_settings()
+    logger.info("Generate index for the provided data")
+
+    # Get the stores and documents or create new ones
+    documents = get_documents()
+    docstore = get_doc_store()
+    vector_store = get_vector_store()
+
+    # Run the ingestion pipeline
+    _ = run_pipeline(docstore, vector_store, documents)
+
+    # Build the index and persist storage
+    persist_storage(docstore, vector_store)
+
+    logger.info("Finished generating the index")
+
+
+if __name__ == "__main__":
+    generate_datasource()

templates/types/streaming/fastapi/app/engine/index.py

@@ -0,0 +1,17 @@
+import logging
+from llama_index.core.indices import VectorStoreIndex
+from app.engine.vectordb import get_vector_store
+
+
+logger = logging.getLogger("uvicorn")
+
+
+def get_index():
+    logger.info("Connecting vector store...")
+    store = get_vector_store()
+    # Load the index from the vector store
+    # If you are using a vector store that doesn't store text,
+    # you must load the index from both the vector store and the document store
+    index = VectorStoreIndex.from_vector_store(store)
+    logger.info("Finished load index from vector store.")
+    return index

templates/types/streaming/fastapi/app/settings.py

@@ -23,8 +23,12 @@ def init_ollama():
     from llama_index.llms.ollama import Ollama
     from llama_index.embeddings.ollama import OllamaEmbedding
 
-    Settings.embed_model = OllamaEmbedding(model_name=os.getenv("EMBEDDING_MODEL"))
-    Settings.llm = Ollama(model=os.getenv("MODEL"))
+    base_url = os.getenv("OLLAMA_BASE_URL") or "http://127.0.0.1:11434"
+    Settings.embed_model = OllamaEmbedding(
+        base_url=base_url,
+        model_name=os.getenv("EMBEDDING_MODEL"),
+    )
+    Settings.llm = Ollama(base_url=base_url, model=os.getenv("MODEL"))
 
 
 def init_openai():

templates/types/streaming/fastapi/main.py

@@ -11,6 +11,7 @@ from fastapi.responses import RedirectResponse
 from app.api.routers.chat import chat_router
 from app.settings import init_settings
 from app.observability import init_observability
+from fastapi.staticfiles import StaticFiles
 
 
 app = FastAPI()
@@ -20,7 +21,6 @@ init_observability()
 
 environment = os.getenv("ENVIRONMENT", "dev")  # Default to 'development' if not set
 
-
 if environment == "dev":
     logger = logging.getLogger("uvicorn")
     logger.warning("Running in development mode - allowing CORS for all origins")
@@ -38,6 +38,8 @@ if environment == "dev":
         return RedirectResponse(url="/docs")
 
 
+if os.path.exists("data"):
+    app.mount("/api/data", StaticFiles(directory="data"), name="static")
 app.include_router(chat_router, prefix="/api/chat")
 
 

templates/types/streaming/fastapi/pyproject.toml

@@ -16,6 +16,7 @@ python-dotenv = "^1.0.0"
 aiostream = "^0.5.2"
 llama-index = "0.10.28"
 llama-index-core = "0.10.28"
+cachetools = "^5.3.3"
 
 [build-system]
 requires = ["poetry-core"]

templates/types/streaming/nextjs/app/api/chat/engine/settings.ts

@@ -56,11 +56,16 @@ function initOpenAI() {
 }
 
 function initOllama() {
+  const config = {
+    host: process.env.OLLAMA_BASE_URL ?? "http://127.0.0.1:11434",
+  };
   Settings.llm = new Ollama({
     model: process.env.MODEL ?? "",
+    config,
   });
   Settings.embedModel = new OllamaEmbedding({
     model: process.env.EMBEDDING_MODEL ?? "",
+    config,
   });
 }
 

templates/types/streaming/nextjs/app/api/data/[path]/route.ts

@@ -0,0 +1,38 @@
+import { readFile } from "fs/promises";
+import { NextRequest, NextResponse } from "next/server";
+import path from "path";
+
+/**
+ * This API is to get file data from ./data folder
+ * It receives path slug and response file data like serve static file
+ */
+export async function GET(
+  _request: NextRequest,
+  { params }: { params: { path: string } },
+) {
+  const slug = params.path;
+
+  if (!slug) {
+    return NextResponse.json({ detail: "Missing file slug" }, { status: 400 });
+  }
+
+  if (slug.includes("..") || path.isAbsolute(slug)) {
+    return NextResponse.json({ detail: "Invalid file path" }, { status: 400 });
+  }
+
+  try {
+    const filePath = path.join(process.cwd(), "data", slug);
+    const blob = await readFile(filePath);
+
+    return new NextResponse(blob, {
+      status: 200,
+      statusText: "OK",
+      headers: {
+        "Content-Length": blob.byteLength.toString(),
+      },
+    });
+  } catch (error) {
+    console.error(error);
+    return NextResponse.json({ detail: "File not found" }, { status: 404 });
+  }
+}

templates/types/streaming/nextjs/app/components/ui/chat/chat-sources.tsx

@@ -1,20 +1,78 @@
-import { ArrowUpRightSquare, Check, Copy } from "lucide-react";
+import { Check, Copy } from "lucide-react";
 import { useMemo } from "react";
 import { Button } from "../button";
 import { HoverCard, HoverCardContent, HoverCardTrigger } from "../hover-card";
+import { getStaticFileDataUrl } from "../lib/url";
 import { SourceData, SourceNode } from "./index";
 import { useCopyToClipboard } from "./use-copy-to-clipboard";
+import PdfDialog from "./widgets/PdfDialog";
 
-const SCORE_THRESHOLD = 0.5;
+const SCORE_THRESHOLD = 0.3;
+
+function SourceNumberButton({ index }: { index: number }) {
+  return (
+    <div className="text-xs w-5 h-5 rounded-full bg-gray-100 mb-2 flex items-center justify-center hover:text-white hover:bg-primary hover:cursor-pointer">
+      {index + 1}
+    </div>
+  );
+}
+
+enum NODE_TYPE {
+  URL,
+  FILE,
+  UNKNOWN,
+}
+
+type NodeInfo = {
+  id: string;
+  type: NODE_TYPE;
+  path?: string;
+  url?: string;
+};
+
+function getNodeInfo(node: SourceNode): NodeInfo {
+  if (typeof node.metadata["URL"] === "string") {
+    const url = node.metadata["URL"];
+    return {
+      id: node.id,
+      type: NODE_TYPE.URL,
+      path: url,
+      url,
+    };
+  }
+  if (typeof node.metadata["file_path"] === "string") {
+    const fileName = node.metadata["file_name"] as string;
+    return {
+      id: node.id,
+      type: NODE_TYPE.FILE,
+      path: node.metadata["file_path"],
+      url: getStaticFileDataUrl(fileName),
+    };
+  }
+
+  return {
+    id: node.id,
+    type: NODE_TYPE.UNKNOWN,
+  };
+}
 
 export function ChatSources({ data }: { data: SourceData }) {
-  const sources = useMemo(() => {
-    return (
-      data.nodes
-        ?.filter((node) => Object.keys(node.metadata).length > 0)
-        ?.filter((node) => (node.score ?? 1) > SCORE_THRESHOLD)
-        .sort((a, b) => (b.score ?? 1) - (a.score ?? 1)) || []
-    );
+  const sources: NodeInfo[] = useMemo(() => {
+    // aggregate nodes by url or file_path (get the highest one by score)
+    const nodesByPath: { [path: string]: NodeInfo } = {};
+
+    data.nodes
+      .filter((node) => (node.score ?? 1) > SCORE_THRESHOLD)
+      .sort((a, b) => (b.score ?? 1) - (a.score ?? 1))
+      .forEach((node) => {
+        const nodeInfo = getNodeInfo(node);
+        const key = nodeInfo.path ?? nodeInfo.id; // use id as key for UNKNOWN type
+        if (!nodesByPath[key]) {
+          nodesByPath[key] = nodeInfo;
+        }
+      });
+
+    return Object.values(nodesByPath);
   }, [data.nodes]);
 
   if (sources.length === 0) return null;
@@ -23,55 +81,52 @@ export function ChatSources({ data }: { data: SourceData }) {
     <div className="space-x-2 text-sm">
       <span className="font-semibold">Sources:</span>
       <div className="inline-flex gap-1 items-center">
-        {sources.map((node: SourceNode, index: number) => (
-          <div key={node.id}>
-            <HoverCard>
-              <HoverCardTrigger>
-                <div className="text-xs w-5 h-5 rounded-full bg-gray-100 mb-2 flex items-center justify-center hover:text-white hover:bg-primary hover:cursor-pointer">
-                  {index + 1}
-                </div>
-              </HoverCardTrigger>
-              <HoverCardContent>
-                <NodeInfo node={node} />
-              </HoverCardContent>
-            </HoverCard>
-          </div>
-        ))}
+        {sources.map((nodeInfo: NodeInfo, index: number) => {
+          if (nodeInfo.path?.endsWith(".pdf")) {
+            return (
+              <PdfDialog
+                key={nodeInfo.id}
+                documentId={nodeInfo.id}
+                url={nodeInfo.url!}
+                path={nodeInfo.path}
+                trigger={<SourceNumberButton index={index} />}
+              />
+            );
+          }
+          return (
+            <div key={nodeInfo.id}>
+              <HoverCard>
+                <HoverCardTrigger>
+                  <SourceNumberButton index={index} />
+                </HoverCardTrigger>
+                <HoverCardContent className="w-[320px]">
+                  <NodeInfo nodeInfo={nodeInfo} />
+                </HoverCardContent>
+              </HoverCard>
+            </div>
+          );
+        })}
       </div>
     </div>
   );
 }
 
-function NodeInfo({ node }: { node: SourceNode }) {
+function NodeInfo({ nodeInfo }: { nodeInfo: NodeInfo }) {
   const { isCopied, copyToClipboard } = useCopyToClipboard({ timeout: 1000 });
 
-  if (typeof node.metadata["URL"] === "string") {
-    // this is a node generated by the web loader, it contains an external URL
-    // add a link to view this URL
+  if (nodeInfo.type !== NODE_TYPE.UNKNOWN) {
+    // this is a node generated by the web loader or file loader,
+    // add a link to view its URL and a button to copy the URL to the clipboard
     return (
-      <a
-        className="space-x-2 flex items-center my-2 hover:text-blue-900"
-        href={node.metadata["URL"]}
-        target="_blank"
-      >
-        <span>{node.metadata["URL"]}</span>
-        <ArrowUpRightSquare className="w-4 h-4" />
-      </a>
-    );
-  }
-
-  if (typeof node.metadata["file_path"] === "string") {
-    // this is a node generated by the file loader, it contains file path
-    // add a button to copy the path to the clipboard
-    const filePath = node.metadata["file_path"];
-    return (
-      <div className="flex items-center px-2 py-1 justify-between my-2">
-        <span>{filePath}</span>
+      <div className="flex items-center my-2">
+        <a className="hover:text-blue-900" href={nodeInfo.url} target="_blank">
+          <span>{nodeInfo.path}</span>
+        </a>
         <Button
-          onClick={() => copyToClipboard(filePath)}
+          onClick={() => copyToClipboard(nodeInfo.path!)}
           size="icon"
           variant="ghost"
-          className="h-12 w-12"
+          className="h-12 w-12 shrink-0"
         >
           {isCopied ? (
             <Check className="h-4 w-4" />
@@ -84,7 +139,6 @@ function NodeInfo({ node }: { node: SourceNode }) {
   }
 
   // node generated by unknown loader, implement renderer by analyzing logged out metadata
-  console.log("Node metadata", node.metadata);
   return (
     <p>
       Sorry, unknown node type. Please add a new renderer in the NodeInfo

templates/types/streaming/nextjs/app/components/ui/chat/markdown.tsx

@@ -1,5 +1,7 @@
+import "katex/dist/katex.min.css";
 import { FC, memo } from "react";
 import ReactMarkdown, { Options } from "react-markdown";
+import rehypeKatex from "rehype-katex";
 import remarkGfm from "remark-gfm";
 import remarkMath from "remark-math";
 
@@ -12,11 +14,27 @@ const MemoizedReactMarkdown: FC<Options> = memo(
     prevProps.className === nextProps.className,
 );
 
+const preprocessLaTeX = (content: string) => {
+  // Replace block-level LaTeX delimiters \[ \] with $$ $$
+  const blockProcessedContent = content.replace(
+    /\\\[(.*?)\\\]/gs,
+    (_, equation) => `$$${equation}$$`,
+  );
+  // Replace inline LaTeX delimiters \( \) with $ $
+  const inlineProcessedContent = blockProcessedContent.replace(
+    /\\\((.*?)\\\)/gs,
+    (_, equation) => `$${equation}$`,
+  );
+  return inlineProcessedContent;
+};
+
 export default function Markdown({ content }: { content: string }) {
+  const processedContent = preprocessLaTeX(content);
   return (
     <MemoizedReactMarkdown
       className="prose dark:prose-invert prose-p:leading-relaxed prose-pre:p-0 break-words custom-markdown"
       remarkPlugins={[remarkGfm, remarkMath]}
+      rehypePlugins={[rehypeKatex as any]}
       components={{
         p({ children }) {
           return <p className="mb-2 last:mb-0">{children}</p>;
@@ -53,7 +71,7 @@ export default function Markdown({ content }: { content: string }) {
         },
       }}
     >
-      {content}
+      {processedContent}
     </MemoizedReactMarkdown>
   );
 }

templates/types/streaming/nextjs/app/components/ui/chat/widgets/PdfDialog.tsx

@@ -0,0 +1,56 @@
+import { PDFViewer, PdfFocusProvider } from "@llamaindex/pdf-viewer";
+import { Button } from "../../button";
+import {
+  Drawer,
+  DrawerClose,
+  DrawerContent,
+  DrawerDescription,
+  DrawerHeader,
+  DrawerTitle,
+  DrawerTrigger,
+} from "../../drawer";
+
+export interface PdfDialogProps {
+  documentId: string;
+  path: string;
+  url: string;
+  trigger: React.ReactNode;
+}
+
+export default function PdfDialog(props: PdfDialogProps) {
+  return (
+    <Drawer direction="left">
+      <DrawerTrigger>{props.trigger}</DrawerTrigger>
+      <DrawerContent className="w-3/5 mt-24 h-full max-h-[96%] ">
+        <DrawerHeader className="flex justify-between">
+          <div className="space-y-2">
+            <DrawerTitle>PDF Content</DrawerTitle>
+            <DrawerDescription>
+              File path:{" "}
+              <a
+                className="hover:text-blue-900"
+                href={props.url}
+                target="_blank"
+              >
+                {props.path}
+              </a>
+            </DrawerDescription>
+          </div>
+          <DrawerClose asChild>
+            <Button variant="outline">Close</Button>
+          </DrawerClose>
+        </DrawerHeader>
+        <div className="m-4">
+          <PdfFocusProvider>
+            <PDFViewer
+              file={{
+                id: props.documentId,
+                url: props.url,
+              }}
+            />
+          </PdfFocusProvider>
+        </div>
+      </DrawerContent>
+    </Drawer>
+  );
+}

templates/types/streaming/nextjs/app/components/ui/drawer.tsx

@@ -0,0 +1,118 @@
+"use client";
+
+import * as React from "react";
+import { Drawer as DrawerPrimitive } from "vaul";
+
+import { cn } from "./lib/utils";
+
+const Drawer = ({
+  shouldScaleBackground = true,
+  ...props
+}: React.ComponentProps<typeof DrawerPrimitive.Root>) => (
+  <DrawerPrimitive.Root
+    shouldScaleBackground={shouldScaleBackground}
+    {...props}
+  />
+);
+Drawer.displayName = "Drawer";
+
+const DrawerTrigger = DrawerPrimitive.Trigger;
+
+const DrawerPortal = DrawerPrimitive.Portal;
+
+const DrawerClose = DrawerPrimitive.Close;
+
+const DrawerOverlay = React.forwardRef<
+  React.ElementRef<typeof DrawerPrimitive.Overlay>,
+  React.ComponentPropsWithoutRef<typeof DrawerPrimitive.Overlay>
+>(({ className, ...props }, ref) => (
+  <DrawerPrimitive.Overlay
+    ref={ref}
+    className={cn("fixed inset-0 z-50 bg-black/80", className)}
+    {...props}
+  />
+));
+DrawerOverlay.displayName = DrawerPrimitive.Overlay.displayName;
+
+const DrawerContent = React.forwardRef<
+  React.ElementRef<typeof DrawerPrimitive.Content>,
+  React.ComponentPropsWithoutRef<typeof DrawerPrimitive.Content>
+>(({ className, children, ...props }, ref) => (
+  <DrawerPortal>
+    <DrawerOverlay />
+    <DrawerPrimitive.Content
+      ref={ref}
+      className={cn(
+        "fixed inset-x-0 bottom-0 z-50 mt-24 flex h-auto flex-col rounded-t-[10px] border bg-background",
+        className,
+      )}
+      {...props}
+    >
+      <div className="mx-auto mt-4 h-2 w-[100px] rounded-full bg-muted" />
+      {children}
+    </DrawerPrimitive.Content>
+  </DrawerPortal>
+));
+DrawerContent.displayName = "DrawerContent";
+
+const DrawerHeader = ({
+  className,
+  ...props
+}: React.HTMLAttributes<HTMLDivElement>) => (
+  <div
+    className={cn("grid gap-1.5 p-4 text-center sm:text-left", className)}
+    {...props}
+  />
+);
+DrawerHeader.displayName = "DrawerHeader";
+
+const DrawerFooter = ({
+  className,
+  ...props
+}: React.HTMLAttributes<HTMLDivElement>) => (
+  <div
+    className={cn("mt-auto flex flex-col gap-2 p-4", className)}
+    {...props}
+  />
+);
+DrawerFooter.displayName = "DrawerFooter";
+
+const DrawerTitle = React.forwardRef<
+  React.ElementRef<typeof DrawerPrimitive.Title>,
+  React.ComponentPropsWithoutRef<typeof DrawerPrimitive.Title>
+>(({ className, ...props }, ref) => (
+  <DrawerPrimitive.Title
+    ref={ref}
+    className={cn(
+      "text-lg font-semibold leading-none tracking-tight",
+      className,
+    )}
+    {...props}
+  />
+));
+DrawerTitle.displayName = DrawerPrimitive.Title.displayName;
+
+const DrawerDescription = React.forwardRef<
+  React.ElementRef<typeof DrawerPrimitive.Description>,
+  React.ComponentPropsWithoutRef<typeof DrawerPrimitive.Description>
+>(({ className, ...props }, ref) => (
+  <DrawerPrimitive.Description
+    ref={ref}
+    className={cn("text-sm text-muted-foreground", className)}
+    {...props}
+  />
+));
+DrawerDescription.displayName = DrawerPrimitive.Description.displayName;
+
+export {
+  Drawer,
+  DrawerClose,
+  DrawerContent,
+  DrawerDescription,
+  DrawerFooter,
+  DrawerHeader,
+  DrawerOverlay,
+  DrawerPortal,
+  DrawerTitle,
+  DrawerTrigger,
+};

templates/types/streaming/nextjs/app/components/ui/lib/url.ts

@@ -0,0 +1,11 @@
+const STORAGE_FOLDER = "data";
+
+export const getStaticFileDataUrl = (filename: string) => {
+  const isUsingBackend = !!process.env.NEXT_PUBLIC_CHAT_API;
+  const fileUrl = `/api/${STORAGE_FOLDER}/${filename}`;
+  if (isUsingBackend) {
+    const backendOrigin = new URL(process.env.NEXT_PUBLIC_CHAT_API!).origin;
+    return `${backendOrigin}/${fileUrl}`;
+  }
+  return fileUrl;
+};

templates/types/streaming/nextjs/package.json

@@ -18,7 +18,7 @@
     "class-variance-authority": "^0.7.0",
     "clsx": "^2.1.1",
     "dotenv": "^16.3.1",
-    "llamaindex": "0.3.9",
+    "llamaindex": "0.3.13",
     "lucide-react": "^0.294.0",
     "next": "^14.0.3",
     "pdf2json": "3.0.5",
@@ -30,8 +30,11 @@
     "remark-code-import": "^1.2.0",
     "remark-gfm": "^3.0.1",
     "remark-math": "^5.1.1",
+    "rehype-katex": "^7.0.0",
     "supports-color": "^8.1.1",
-    "tailwind-merge": "^2.1.0"
+    "tailwind-merge": "^2.1.0",
+    "vaul": "^0.9.1",
+    "@llamaindex/pdf-viewer": "^1.1.1"
   },
   "devDependencies": {
     "@types/node": "^20.10.3",

tsconfig.json

@@ -11,14 +11,16 @@
     "forceConsistentCasingInFileNames": true,
     "incremental": true,
     "outDir": "./lib",
-    "tsBuildInfoFile": "./lib/.tsbuildinfo"
+    "tsBuildInfoFile": "./lib/.tsbuildinfo",
+    "typeRoots": ["./types", "./node_modules/@types"]
   },
   "include": [
     "create-app.ts",
     "index.ts",
     "./helpers",
     "questions.ts",
-    "package.json"
+    "package.json",
+    "types/**/*"
   ],
   "exclude": ["dist"]
 }

types/global-agent.d.ts

@@ -0,0 +1 @@
+declare module "global-agent/bootstrap";