Starting April 29, 2025, Gemini 1.5 Pro and Gemini 1.5 Flash models are not available in projects that have no prior usage of these models, including new projects. For details, see
Model versions and lifecycle.
Use the quickstart to get familiar with RAG
Stay organized with collections
Save and categorize content based on your preferences.
This quickstart demonstrates how to use the RAG API.
Explore further
For detailed documentation that includes this code sample, see the following:
Code sample
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],[],[],[],null,["# Use the quickstart to get familiar with RAG\n\nThis quickstart demonstrates how to use the RAG API.\n\nExplore further\n---------------\n\n\nFor detailed documentation that includes this code sample, see the following:\n\n- [RAG quickstart for Python](/vertex-ai/generative-ai/docs/rag-engine/rag-quickstart)\n\nCode sample\n-----------\n\n### Python\n\n\nBefore trying this sample, follow the Python setup instructions in the\n[Vertex AI quickstart using\nclient libraries](/vertex-ai/docs/start/client-libraries).\n\n\nFor more information, see the\n[Vertex AI Python API\nreference documentation](/python/docs/reference/aiplatform/latest).\n\n\nTo authenticate to Vertex AI, set up Application Default Credentials.\nFor more information, see\n\n[Set up authentication for a local development environment](/docs/authentication/set-up-adc-local-dev-environment).\n\n from vertexai import rag\n from vertexai.generative_models import https://cloud.google.com/python/docs/reference/vertexai/latest/vertexai.preview.generative_models.GenerativeModel.html, https://cloud.google.com/python/docs/reference/vertexai/latest/vertexai.preview.generative_models.Tool.html\n import https://cloud.google.com/python/docs/reference/vertexai/latest/\n\n # Create a RAG Corpus, Import Files, and Generate a response\n\n # TODO(developer): Update and un-comment below lines\n # PROJECT_ID = \"your-project-id\"\n # display_name = \"test_corpus\"\n # paths = [\"https://drive.google.com/file/d/123\", \"gs://my_bucket/my_files_dir\"] # Supports Google Cloud Storage and Google Drive Links\n\n # Initialize Vertex AI API once per session\n https://cloud.google.com/python/docs/reference/vertexai/latest/.init(project=PROJECT_ID, location=\"us-central1\")\n\n # Create RagCorpus\n # Configure embedding model, for example \"text-embedding-005\".\n embedding_model_config = rag.RagEmbeddingModelConfig(\n vertex_prediction_endpoint=rag.VertexPredictionEndpoint(\n publisher_model=\"publishers/google/models/text-embedding-005\"\n )\n )\n\n rag_corpus = rag.create_corpus(\n display_name=display_name,\n backend_config=rag.RagVectorDbConfig(\n rag_embedding_model_config=embedding_model_config\n ),\n )\n\n # Import Files to the RagCorpus\n rag.import_files(\n rag_corpus.name,\n paths,\n # Optional\n transformation_config=rag.TransformationConfig(\n chunking_config=rag.ChunkingConfig(\n chunk_size=512,\n chunk_overlap=100,\n ),\n ),\n max_embedding_requests_per_min=1000, # Optional\n )\n\n # Direct context retrieval\n rag_retrieval_config = rag.RagRetrievalConfig(\n top_k=3, # Optional\n filter=rag.Filter(vector_distance_threshold=0.5), # Optional\n )\n response = rag.retrieval_query(\n rag_resources=[\n rag.RagResource(\n rag_corpus=rag_corpus.name,\n # Optional: supply IDs from `rag.list_files()`.\n # rag_file_ids=[\"rag-file-1\", \"rag-file-2\", ...],\n )\n ],\n text=\"What is RAG and why it is helpful?\",\n rag_retrieval_config=rag_retrieval_config,\n )\n print(response)\n\n # Enhance generation\n # Create a RAG retrieval tool\n rag_retrieval_tool = https://cloud.google.com/python/docs/reference/vertexai/latest/vertexai.preview.generative_models.Tool.html.from_retrieval(\n retrieval=rag.Retrieval(\n source=rag.VertexRagStore(\n rag_resources=[\n rag.RagResource(\n rag_corpus=rag_corpus.name, # Currently only 1 corpus is allowed.\n # Optional: supply IDs from `rag.list_files()`.\n # rag_file_ids=[\"rag-file-1\", \"rag-file-2\", ...],\n )\n ],\n rag_retrieval_config=rag_retrieval_config,\n ),\n )\n )\n\n # Create a Gemini model instance\n rag_model = GenerativeModel(\n model_name=\"gemini-2.0-flash-001\", tools=[rag_retrieval_tool]\n )\n\n # Generate response\n response = rag_model.https://cloud.google.com/python/docs/reference/vertexai/latest/vertexai.preview.generative_models.GenerativeModel.html#vertexai_preview_generative_models_GenerativeModel_generate_content(\"What is RAG and why it is helpful?\")\n print(response.text)\n # Example response:\n # RAG stands for Retrieval-Augmented Generation.\n # It's a technique used in AI to enhance the quality of responses\n # ...\n\nWhat's next\n-----------\n\n\nTo search and filter code samples for other Google Cloud products, see the\n[Google Cloud sample browser](/docs/samples?product=generativeaionvertexai)."]]