<\/span>Prerequisites<\/span><\/h2>\n\n\n\nBefore we begin, ensure you have the following:<\/p>\n\n\n\n
\n- Python installed on your machine.<\/li>\n\n\n\n
- Access to Jina.AI API and TiDB serverless instance.\n
\n- Please make sure you have created a TiDB Serverless cluster with vector support enabled. Join the waitlist for the private beta at tidb.cloud\/ai<\/a>.\n
\n- Sign up TiDB Cloud<\/a><\/li>\n\n\n\n
- Follow this tutorial<\/a> to create a TiDB Serverless cluster with vector support enabled<\/li>\n\n\n\n
- Navigate to the Clusters<\/a> page, and then click the name of your target cluster to go to its overview page<\/li>\n\n\n\n
- Click Connect in the upper-right corner.<\/li>\n\n\n\n
- In the connection dialog, select General from the Connect With dropdown and keep the default setting of the Endpoint Type as Public.<\/li>\n\n\n\n
- If you have not set a password yet, click Create password to generate a random password.<\/li>\n\n\n\n
- Save the connection parameters to a safe place. You will need them to connect to the TiDB Serverless cluster in the following steps.<\/li>\n<\/ul>\n<\/li>\n\n\n\n
- Generate an API token for Jina.AI embeddings API\n
\n- Go to https:\/\/jina.ai\/embeddings\/<\/a><\/li>\n\n\n\n
- Scroll down to the API part, you will get a token for free with 1million tokens:<\/li>\n\n\n\n
- Save this token for future use. If you exceed the free quota of 1 million tokens, you will need this token to charge for additional usage.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n\n
![\"Embedding](\"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png\")
<\/figure>\n<\/div>\n\n\n\n- Required Python libraries installed (
requests<\/code>, sqlalchemy<\/code>, pymysql<\/code>, dotenv<\/code>).<\/li>\n<\/ul>\n\n\n\n<\/span>Step 1: Setting Up Your Environment<\/span><\/h2>\n\n\n\nFirst, let’s set up our environment by loading the necessary credentials and libraries.<\/p>\n\n\n\n
import os\nimport requests\nimport dotenv\nfrom sqlalchemy import Column, Integer, String, create_engine, URL\nfrom sqlalchemy.orm import Session, declarative_base\nfrom tidb_vector.sqlalchemy import VectorType\n\ndotenv.load_dotenv()\n\nJINAAI_API_KEY = os.getenv('JINAAI_API_KEY')\nTIDB_USERNAME = os.getenv('TIDB_USERNAME')\nTIDB_PASSWORD = os.getenv('TIDB_PASSWORD')\nTIDB_HOST = os.getenv('TIDB_HOST')\nTIDB_PORT = os.getenv('TIDB_PORT')\nTIDB_DATABASE = os.getenv('TIDB_DATABASE')<\/code><\/pre>\n\n\n\n<\/span>Step 2: Generating Embeddings with Jina.AI<\/span><\/h2>\n\n\n\nWe will generate text embeddings using the Jina.AI API. The embeddings will be stored in TiDB for efficient querying.<\/p>\n\n\n\n
Define the Function to Generate Embeddings<\/strong><\/p>\n\n\n\ndef generate_embeddings(text: str):\n JINAAI_API_URL = 'https:\/\/api.jina.ai\/v1\/embeddings'\n JINAAI_HEADERS = {\n 'Content-Type': 'application\/json',\n 'Authorization': f'Bearer {JINAAI_API_KEY}'\n }\n JINAAI_REQUEST_DATA = {\n 'input': [text],\n 'model': 'jina-embeddings-v2-base-en' # Dimensions: 768\n }\n response = requests.post(JINAAI_API_URL, headers=JINAAI_HEADERS, json=JINAAI_REQUEST_DATA)\n return response.json()['data'][0]['embedding']\n\nTEXTS = [\n 'Jina AI offers best-in-class embeddings, reranker and prompt optimizer, enabling advanced multimodal AI.',\n 'TiDB is an open-source MySQL-compatible database that supports Hybrid Transactional and Analytical Processing (HTAP) workloads.',\n]\n\ndata = []\nfor text in TEXTS:\n embedding = generate_embeddings(text)\n data.append({\n 'text': text,\n 'embedding': embedding\n })<\/code><\/pre>\n\n\n\n<\/span>Step 3: Connecting to TiDB and Creating a Vector-Enabled Table<\/span><\/h2>\n\n\n\nNext, we’ll connect to TiDB and create a table that can store text data and their corresponding embeddings.<\/p>\n\n\n\n
Connect to TiDB and Define the Table<\/strong><\/p>\n\n\n\nurl = URL(\n drivername=\"mysql+pymysql\",\n username=TIDB_USERNAME,\n password=TIDB_PASSWORD,\n host=TIDB_HOST,\n port=int(TIDB_PORT),\n database=TIDB_DATABASE,\n query={\"ssl_verify_cert\": True, \"ssl_verify_identity\": True},\n)\nengine = create_engine(url, pool_recycle=300)\nBase = declarative_base()\n\nclass Document(Base):\n __tablename__ = \"jinaai_tidb_demo_documents\"\n\n id = Column(Integer, primary_key=True)\n content = Column(String(255), nullable=False)\n content_vec = Column(\n VectorType(dim=768),\n comment=\"hnsw(distance=l2)\"\n )\nBase.metadata.create_all(engine)<\/code><\/pre>\n\n\n\n<\/span>Step 4: Inserting Data into TiDB<\/span><\/h2>\n\n\n\nNow, we will insert the generated embeddings and their corresponding texts into the TiDB table.<\/p>\n\n\n\n
Insert Data<\/strong><\/p>\n\n\n\nwith Session(engine) as session:\n print('- Inserting Data to TiDB...')\n for item in data:\n print(f' - Inserting: {item[\"text\"]}')\n session.add(Document(\n content=item['text'],\n content_vec=item['embedding']\n ))\n session.commit()<\/code><\/pre>\n\n\n\n<\/span>Step 5: Querying Data from TiDB<\/span><\/h2>\n\n\n\nFinally, let’s query the data from TiDB to find the most relevant document based on a sample query.<\/p>\n\n\n\n
Query the Data<\/strong><\/p>\n\n\n\nquery = 'What is TiDB?'\nquery_embedding = generate_embeddings(query)\nwith Session(engine) as session:\n print('- List All Documents and Their Distances to the Query:')\n for doc, distance in session.query(\n Document,\n Document.content_vec.cosine_distance(query_embedding).label('distance')\n ).all():\n print(f' - {doc.content}: {distance}')\n\n print('- The Most Relevant Document and Its Distance to the Query:')\n doc, distance = session.query(\n Document,\n Document.content_vec.cosine_distance(query_embedding).label('distance')\n ).order_by(\n 'distance'\n ).limit(1).first()\n print(f' - {doc.content}: {distance}')<\/code><\/pre>\n\n\n\nOutput<\/strong><\/p>\n\n\n\nWhen you run the above script, you should see output similar to this:<\/p>\n\n\n\n
- Inserting Data to TiDB...\n - Inserting: Jina AI offers best-in-class embeddings, reranker and prompt optimizer, enabling advanced multimodal AI.\n - Inserting: TiDB is an open-source MySQL-compatible database that supports Hybrid Transactional and Analytical Processing (HTAP) workloads.\n- List All Documents and Their Distances to the Query:\n - Jina AI offers best-in-class embeddings, reranker and prompt optimizer, enabling advanced multimodal AI.: 0.3585317326132522\n - TiDB is an open-source MySQL-compatible database that supports Hybrid Transactional and Analytical Processing (HTAP) workloads.: 0.10858658947444844\n- The Most Relevant Document and Its Distance to the Query:\n - TiDB is an open-source MySQL-compatible database that supports Hybrid Transactional and Analytical Processing (HTAP) workloads.: 0.10858658947444844<\/code><\/pre>\n\n\n\n<\/span>Conclusion<\/span><\/h2>\n\n\n\nCongratulations! You have successfully built a RAG system using Jina.AI embeddings API and TiDB’s Vector Storage. This setup allows you to efficiently store and query vectorized data, enabling advanced search capabilities in your applications.<\/p>\n\n\n\n
Feel free to expand this tutorial by adding more functionalities, such as handling larger datasets, optimizing queries, and integrating with other machine learning models.<\/p>\n\n\n\n
Full demo: https:\/\/github.com\/pingcap\/tidb-vector-python\/tree\/main\/examples\/jina-ai-embeddings-demo<\/a><\/p>\n\n\n\nThere are also more demos to demonstrate how to build AI apps, how to use TiDB as vector storage:<\/p>\n\n\n\n
\n- OpenAI Embedding<\/a>: use the OpenAI embedding model to generate vectors for text data.<\/li>\n\n\n\n
- Image Search<\/a>: use the OpenAI CLIP model to generate vectors for image and text.<\/li>\n\n\n\n
- LlamaIndex RAG with UI<\/a>: use the LlamaIndex to build an RAG(Retrieval-Augmented Generation)<\/a> application.<\/li>\n\n\n\n
- Chat with URL<\/a>: use LlamaIndex to build an RAG(Retrieval-Augmented Generation)<\/a> application that can chat with a URL.<\/li>\n\n\n\n
- GraphRAG<\/a>: 20 lines code of using TiDB Serverless to build a Knowledge Graph based RAG application.<\/li>\n\n\n\n
- GraphRAG Step by Step Tutorial<\/a>: Step by step tutorial to build a Knowledge Graph based RAG application with Colab notebook. In this tutorial, you will learn how to extract knowledge from a text corpus, build a Knowledge Graph, store the Knowledge Graph in TiDB Serverless, and search from the Knowledge Graph.<\/li>\n\n\n\n
- Vector Search Notebook with SQLAlchemy<\/a>: use SQLAlchemy<\/a> to interact with TiDB Serverless: connect db, index&store data and then search vectors.<\/li>\n<\/ul>","protected":false},"excerpt":{"rendered":"
In this tutorial, we will walk through building a Retrieval-Augmented Generation (RAG) system using the Jina.AI embeddings API and TiDB’s Vector Storage. This combination allows for efficient handling of vectorized data and powerful retrieval capabilities. We will cover the following steps: Prerequisites Before we begin, ensure you have the following: Step 1: Setting Up Your […]<\/p>\n","protected":false},"author":8,"featured_media":0,"template":"","class_list":["post-17768","article","type-article","status-publish","hentry"],"acf":[],"yoast_head":"\n
Build RAG with Jina.AI Embeddings API and TiDB Vector Storage<\/title>\n<meta name=\"description\" content=\"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills\" \/>\n<meta name=\"robots\" content=\"noindex, follow\" \/>\n<meta property=\"og:locale\" content=\"ko_KR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage\" \/>\n<meta property=\"og:description\" content=\"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.pingcap.com\/ko\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\" \/>\n<meta property=\"og:site_name\" content=\"TiDB\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/facebook.com\/pingcap2015\" \/>\n<meta property=\"article:modified_time\" content=\"2024-06-23T15:33:57+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:site\" content=\"@PingCAP\" \/>\n<meta name=\"twitter:label1\" content=\"\uc608\uc0c1 \ub418\ub294 \ud310\ub3c5 \uc2dc\uac04\" \/>\n\t<meta name=\"twitter:data1\" content=\"6\ubd84\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\",\"url\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\",\"name\":\"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage\",\"isPartOf\":{\"@id\":\"https:\/\/www.pingcap.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png\",\"datePublished\":\"2024-06-23T15:33:54+00:00\",\"dateModified\":\"2024-06-23T15:33:57+00:00\",\"description\":\"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills\",\"breadcrumb\":{\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#breadcrumb\"},\"inLanguage\":\"ko-KR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage\",\"url\":\"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png\",\"contentUrl\":\"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.pingcap.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Articles\",\"item\":\"https:\/\/www.pingcap.com\/article\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.pingcap.com\/#website\",\"url\":\"https:\/\/www.pingcap.com\/\",\"name\":\"TiDB\",\"description\":\"TiDB | SQL at Scale\",\"publisher\":{\"@id\":\"https:\/\/www.pingcap.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.pingcap.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ko-KR\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.pingcap.com\/#organization\",\"name\":\"PingCAP\",\"url\":\"https:\/\/www.pingcap.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png\",\"contentUrl\":\"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png\",\"width\":811,\"height\":232,\"caption\":\"PingCAP\"},\"image\":{\"@id\":\"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/facebook.com\/pingcap2015\",\"https:\/\/x.com\/PingCAP\",\"https:\/\/linkedin.com\/company\/pingcap\",\"https:\/\/youtube.com\/channel\/UCuq4puT32DzHKT5rU1IZpIA\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage","description":"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills","robots":{"index":"noindex","follow":"follow"},"og_locale":"ko_KR","og_type":"article","og_title":"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage","og_description":"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills","og_url":"https:\/\/www.pingcap.com\/ko\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/","og_site_name":"TiDB","article_publisher":"https:\/\/facebook.com\/pingcap2015","article_modified_time":"2024-06-23T15:33:57+00:00","og_image":[{"url":"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png","type":"","width":"","height":""}],"twitter_card":"summary_large_image","twitter_site":"@PingCAP","twitter_misc":{"\uc608\uc0c1 \ub418\ub294 \ud310\ub3c5 \uc2dc\uac04":"6\ubd84"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/","url":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/","name":"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage","isPartOf":{"@id":"https:\/\/www.pingcap.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage"},"image":{"@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage"},"thumbnailUrl":"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png","datePublished":"2024-06-23T15:33:54+00:00","dateModified":"2024-06-23T15:33:57+00:00","description":"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills","breadcrumb":{"@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#breadcrumb"},"inLanguage":"ko-KR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/"]}]},{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage","url":"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png","contentUrl":"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png"},{"@type":"BreadcrumbList","@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.pingcap.com\/"},{"@type":"ListItem","position":2,"name":"Articles","item":"https:\/\/www.pingcap.com\/article\/"},{"@type":"ListItem","position":3,"name":"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage"}]},{"@type":"WebSite","@id":"https:\/\/www.pingcap.com\/#website","url":"https:\/\/www.pingcap.com\/","name":"\ud2f0DB","description":"TiDB | SQL at Scale","publisher":{"@id":"https:\/\/www.pingcap.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.pingcap.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ko-KR"},{"@type":"Organization","@id":"https:\/\/www.pingcap.com\/#organization","name":"PingCAP","url":"https:\/\/www.pingcap.com\/","logo":{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/","url":"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png","contentUrl":"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png","width":811,"height":232,"caption":"PingCAP"},"image":{"@id":"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/facebook.com\/pingcap2015","https:\/\/x.com\/PingCAP","https:\/\/linkedin.com\/company\/pingcap","https:\/\/youtube.com\/channel\/UCuq4puT32DzHKT5rU1IZpIA"]}]}},"card_markup":" <a class=\"card-article\" href=\"https:\/\/www.pingcap.com\/ko\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\"> <h3>Build RAG with Jina.AI Embeddings API and TiDB Vector Storage<\/h3> <p>In this tutorial, we will walk through building a Retrieval-Augmented Generation (RAG) system using the Jina.AI embeddings API and TiDB’s Vector Storage. This combination allows for efficient handling of vectorized data and powerful retrieval capabilities. We will cover the following steps: Prerequisites Before we begin, ensure you have the following: Step 1: Setting Up Your […]<\/p> <\/a>","_links":{"self":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/article\/17768","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/article"}],"about":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/types\/article"}],"author":[{"embeddable":true,"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/users\/8"}],"wp:attachment":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/media?parent=17768"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
- \n
- Please make sure you have created a TiDB Serverless cluster with vector support enabled. Join the waitlist for the private beta at tidb.cloud\/ai<\/a>.\n
- \n
- Sign up TiDB Cloud<\/a><\/li>\n\n\n\n
- Follow this tutorial<\/a> to create a TiDB Serverless cluster with vector support enabled<\/li>\n\n\n\n
- Navigate to the Clusters<\/a> page, and then click the name of your target cluster to go to its overview page<\/li>\n\n\n\n
- Click Connect in the upper-right corner.<\/li>\n\n\n\n
- In the connection dialog, select General from the Connect With dropdown and keep the default setting of the Endpoint Type as Public.<\/li>\n\n\n\n
- If you have not set a password yet, click Create password to generate a random password.<\/li>\n\n\n\n
- Save the connection parameters to a safe place. You will need them to connect to the TiDB Serverless cluster in the following steps.<\/li>\n<\/ul>\n<\/li>\n\n\n\n
- Generate an API token for Jina.AI embeddings API\n
- \n
- Go to https:\/\/jina.ai\/embeddings\/<\/a><\/li>\n\n\n\n
- Scroll down to the API part, you will get a token for free with 1million tokens:<\/li>\n\n\n\n
- Save this token for future use. If you exceed the free quota of 1 million tokens, you will need this token to charge for additional usage.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n
\n<\/figure>\n<\/div>\n\n\n- \n
- Required Python libraries installed (
requests<\/code>,sqlalchemy<\/code>,pymysql<\/code>,dotenv<\/code>).<\/li>\n<\/ul>\n\n\n\n<\/span>Step 1: Setting Up Your Environment<\/span><\/h2>\n\n\n\n
First, let’s set up our environment by loading the necessary credentials and libraries.<\/p>\n\n\n\n
import os\nimport requests\nimport dotenv\nfrom sqlalchemy import Column, Integer, String, create_engine, URL\nfrom sqlalchemy.orm import Session, declarative_base\nfrom tidb_vector.sqlalchemy import VectorType\n\ndotenv.load_dotenv()\n\nJINAAI_API_KEY = os.getenv('JINAAI_API_KEY')\nTIDB_USERNAME = os.getenv('TIDB_USERNAME')\nTIDB_PASSWORD = os.getenv('TIDB_PASSWORD')\nTIDB_HOST = os.getenv('TIDB_HOST')\nTIDB_PORT = os.getenv('TIDB_PORT')\nTIDB_DATABASE = os.getenv('TIDB_DATABASE')<\/code><\/pre>\n\n\n\n<\/span>Step 2: Generating Embeddings with Jina.AI<\/span><\/h2>\n\n\n\n
We will generate text embeddings using the Jina.AI API. The embeddings will be stored in TiDB for efficient querying.<\/p>\n\n\n\n
Define the Function to Generate Embeddings<\/strong><\/p>\n\n\n\n
def generate_embeddings(text: str):\n JINAAI_API_URL = 'https:\/\/api.jina.ai\/v1\/embeddings'\n JINAAI_HEADERS = {\n 'Content-Type': 'application\/json',\n 'Authorization': f'Bearer {JINAAI_API_KEY}'\n }\n JINAAI_REQUEST_DATA = {\n 'input': [text],\n 'model': 'jina-embeddings-v2-base-en' # Dimensions: 768\n }\n response = requests.post(JINAAI_API_URL, headers=JINAAI_HEADERS, json=JINAAI_REQUEST_DATA)\n return response.json()['data'][0]['embedding']\n\nTEXTS = [\n 'Jina AI offers best-in-class embeddings, reranker and prompt optimizer, enabling advanced multimodal AI.',\n 'TiDB is an open-source MySQL-compatible database that supports Hybrid Transactional and Analytical Processing (HTAP) workloads.',\n]\n\ndata = []\nfor text in TEXTS:\n embedding = generate_embeddings(text)\n data.append({\n 'text': text,\n 'embedding': embedding\n })<\/code><\/pre>\n\n\n\n<\/span>Step 3: Connecting to TiDB and Creating a Vector-Enabled Table<\/span><\/h2>\n\n\n\n
Next, we’ll connect to TiDB and create a table that can store text data and their corresponding embeddings.<\/p>\n\n\n\n
Connect to TiDB and Define the Table<\/strong><\/p>\n\n\n\n
url = URL(\n drivername=\"mysql+pymysql\",\n username=TIDB_USERNAME,\n password=TIDB_PASSWORD,\n host=TIDB_HOST,\n port=int(TIDB_PORT),\n database=TIDB_DATABASE,\n query={\"ssl_verify_cert\": True, \"ssl_verify_identity\": True},\n)\nengine = create_engine(url, pool_recycle=300)\nBase = declarative_base()\n\nclass Document(Base):\n __tablename__ = \"jinaai_tidb_demo_documents\"\n\n id = Column(Integer, primary_key=True)\n content = Column(String(255), nullable=False)\n content_vec = Column(\n VectorType(dim=768),\n comment=\"hnsw(distance=l2)\"\n )\nBase.metadata.create_all(engine)<\/code><\/pre>\n\n\n\n<\/span>Step 4: Inserting Data into TiDB<\/span><\/h2>\n\n\n\n
Now, we will insert the generated embeddings and their corresponding texts into the TiDB table.<\/p>\n\n\n\n
Insert Data<\/strong><\/p>\n\n\n\n
with Session(engine) as session:\n print('- Inserting Data to TiDB...')\n for item in data:\n print(f' - Inserting: {item[\"text\"]}')\n session.add(Document(\n content=item['text'],\n content_vec=item['embedding']\n ))\n session.commit()<\/code><\/pre>\n\n\n\n<\/span>Step 5: Querying Data from TiDB<\/span><\/h2>\n\n\n\n
Finally, let’s query the data from TiDB to find the most relevant document based on a sample query.<\/p>\n\n\n\n
Query the Data<\/strong><\/p>\n\n\n\n
query = 'What is TiDB?'\nquery_embedding = generate_embeddings(query)\nwith Session(engine) as session:\n print('- List All Documents and Their Distances to the Query:')\n for doc, distance in session.query(\n Document,\n Document.content_vec.cosine_distance(query_embedding).label('distance')\n ).all():\n print(f' - {doc.content}: {distance}')\n\n print('- The Most Relevant Document and Its Distance to the Query:')\n doc, distance = session.query(\n Document,\n Document.content_vec.cosine_distance(query_embedding).label('distance')\n ).order_by(\n 'distance'\n ).limit(1).first()\n print(f' - {doc.content}: {distance}')<\/code><\/pre>\n\n\n\nOutput<\/strong><\/p>\n\n\n\n
When you run the above script, you should see output similar to this:<\/p>\n\n\n\n
- Inserting Data to TiDB...\n - Inserting: Jina AI offers best-in-class embeddings, reranker and prompt optimizer, enabling advanced multimodal AI.\n - Inserting: TiDB is an open-source MySQL-compatible database that supports Hybrid Transactional and Analytical Processing (HTAP) workloads.\n- List All Documents and Their Distances to the Query:\n - Jina AI offers best-in-class embeddings, reranker and prompt optimizer, enabling advanced multimodal AI.: 0.3585317326132522\n - TiDB is an open-source MySQL-compatible database that supports Hybrid Transactional and Analytical Processing (HTAP) workloads.: 0.10858658947444844\n- The Most Relevant Document and Its Distance to the Query:\n - TiDB is an open-source MySQL-compatible database that supports Hybrid Transactional and Analytical Processing (HTAP) workloads.: 0.10858658947444844<\/code><\/pre>\n\n\n\n<\/span>Conclusion<\/span><\/h2>\n\n\n\n
Congratulations! You have successfully built a RAG system using Jina.AI embeddings API and TiDB’s Vector Storage. This setup allows you to efficiently store and query vectorized data, enabling advanced search capabilities in your applications.<\/p>\n\n\n\n
Feel free to expand this tutorial by adding more functionalities, such as handling larger datasets, optimizing queries, and integrating with other machine learning models.<\/p>\n\n\n\n
Full demo: https:\/\/github.com\/pingcap\/tidb-vector-python\/tree\/main\/examples\/jina-ai-embeddings-demo<\/a><\/p>\n\n\n\n
There are also more demos to demonstrate how to build AI apps, how to use TiDB as vector storage:<\/p>\n\n\n\n
- \n
- OpenAI Embedding<\/a>: use the OpenAI embedding model to generate vectors for text data.<\/li>\n\n\n\n
- Image Search<\/a>: use the OpenAI CLIP model to generate vectors for image and text.<\/li>\n\n\n\n
- LlamaIndex RAG with UI<\/a>: use the LlamaIndex to build an RAG(Retrieval-Augmented Generation)<\/a> application.<\/li>\n\n\n\n
- Chat with URL<\/a>: use LlamaIndex to build an RAG(Retrieval-Augmented Generation)<\/a> application that can chat with a URL.<\/li>\n\n\n\n
- GraphRAG<\/a>: 20 lines code of using TiDB Serverless to build a Knowledge Graph based RAG application.<\/li>\n\n\n\n
- GraphRAG Step by Step Tutorial<\/a>: Step by step tutorial to build a Knowledge Graph based RAG application with Colab notebook. In this tutorial, you will learn how to extract knowledge from a text corpus, build a Knowledge Graph, store the Knowledge Graph in TiDB Serverless, and search from the Knowledge Graph.<\/li>\n\n\n\n
- Vector Search Notebook with SQLAlchemy<\/a>: use SQLAlchemy<\/a> to interact with TiDB Serverless: connect db, index&store data and then search vectors.<\/li>\n<\/ul>","protected":false},"excerpt":{"rendered":"
In this tutorial, we will walk through building a Retrieval-Augmented Generation (RAG) system using the Jina.AI embeddings API and TiDB’s Vector Storage. This combination allows for efficient handling of vectorized data and powerful retrieval capabilities. We will cover the following steps: Prerequisites Before we begin, ensure you have the following: Step 1: Setting Up Your […]<\/p>\n","protected":false},"author":8,"featured_media":0,"template":"","class_list":["post-17768","article","type-article","status-publish","hentry"],"acf":[],"yoast_head":"\n
Build RAG with Jina.AI Embeddings API and TiDB Vector Storage<\/title>\n<meta name=\"description\" content=\"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills\" \/>\n<meta name=\"robots\" content=\"noindex, follow\" \/>\n<meta property=\"og:locale\" content=\"ko_KR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage\" \/>\n<meta property=\"og:description\" content=\"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.pingcap.com\/ko\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\" \/>\n<meta property=\"og:site_name\" content=\"TiDB\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/facebook.com\/pingcap2015\" \/>\n<meta property=\"article:modified_time\" content=\"2024-06-23T15:33:57+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:site\" content=\"@PingCAP\" \/>\n<meta name=\"twitter:label1\" content=\"\uc608\uc0c1 \ub418\ub294 \ud310\ub3c5 \uc2dc\uac04\" \/>\n\t<meta name=\"twitter:data1\" content=\"6\ubd84\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\",\"url\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\",\"name\":\"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage\",\"isPartOf\":{\"@id\":\"https:\/\/www.pingcap.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png\",\"datePublished\":\"2024-06-23T15:33:54+00:00\",\"dateModified\":\"2024-06-23T15:33:57+00:00\",\"description\":\"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills\",\"breadcrumb\":{\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#breadcrumb\"},\"inLanguage\":\"ko-KR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage\",\"url\":\"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png\",\"contentUrl\":\"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.pingcap.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Articles\",\"item\":\"https:\/\/www.pingcap.com\/article\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.pingcap.com\/#website\",\"url\":\"https:\/\/www.pingcap.com\/\",\"name\":\"TiDB\",\"description\":\"TiDB | SQL at Scale\",\"publisher\":{\"@id\":\"https:\/\/www.pingcap.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.pingcap.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ko-KR\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.pingcap.com\/#organization\",\"name\":\"PingCAP\",\"url\":\"https:\/\/www.pingcap.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png\",\"contentUrl\":\"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png\",\"width\":811,\"height\":232,\"caption\":\"PingCAP\"},\"image\":{\"@id\":\"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/facebook.com\/pingcap2015\",\"https:\/\/x.com\/PingCAP\",\"https:\/\/linkedin.com\/company\/pingcap\",\"https:\/\/youtube.com\/channel\/UCuq4puT32DzHKT5rU1IZpIA\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage","description":"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills","robots":{"index":"noindex","follow":"follow"},"og_locale":"ko_KR","og_type":"article","og_title":"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage","og_description":"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills","og_url":"https:\/\/www.pingcap.com\/ko\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/","og_site_name":"TiDB","article_publisher":"https:\/\/facebook.com\/pingcap2015","article_modified_time":"2024-06-23T15:33:57+00:00","og_image":[{"url":"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png","type":"","width":"","height":""}],"twitter_card":"summary_large_image","twitter_site":"@PingCAP","twitter_misc":{"\uc608\uc0c1 \ub418\ub294 \ud310\ub3c5 \uc2dc\uac04":"6\ubd84"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/","url":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/","name":"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage","isPartOf":{"@id":"https:\/\/www.pingcap.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage"},"image":{"@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage"},"thumbnailUrl":"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png","datePublished":"2024-06-23T15:33:54+00:00","dateModified":"2024-06-23T15:33:57+00:00","description":"In this tutorial, we will build a RAG system using the Jina.AI embeddings API and TiDB's Vector Storage and SQLAlchemy with MySQL skills","breadcrumb":{"@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#breadcrumb"},"inLanguage":"ko-KR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/"]}]},{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#primaryimage","url":"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png","contentUrl":"https:\/\/static.pingcap.com\/files\/2024\/06\/23083113\/Embedding-API.png"},{"@type":"BreadcrumbList","@id":"https:\/\/www.pingcap.com\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.pingcap.com\/"},{"@type":"ListItem","position":2,"name":"Articles","item":"https:\/\/www.pingcap.com\/article\/"},{"@type":"ListItem","position":3,"name":"Build RAG with Jina.AI Embeddings API and TiDB Vector Storage"}]},{"@type":"WebSite","@id":"https:\/\/www.pingcap.com\/#website","url":"https:\/\/www.pingcap.com\/","name":"\ud2f0DB","description":"TiDB | SQL at Scale","publisher":{"@id":"https:\/\/www.pingcap.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.pingcap.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ko-KR"},{"@type":"Organization","@id":"https:\/\/www.pingcap.com\/#organization","name":"PingCAP","url":"https:\/\/www.pingcap.com\/","logo":{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/","url":"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png","contentUrl":"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png","width":811,"height":232,"caption":"PingCAP"},"image":{"@id":"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/facebook.com\/pingcap2015","https:\/\/x.com\/PingCAP","https:\/\/linkedin.com\/company\/pingcap","https:\/\/youtube.com\/channel\/UCuq4puT32DzHKT5rU1IZpIA"]}]}},"card_markup":" <a class=\"card-article\" href=\"https:\/\/www.pingcap.com\/ko\/article\/build-rag-with-jina-ai-embeddings-api-and-tidb-vector-storage\/\"> <h3>Build RAG with Jina.AI Embeddings API and TiDB Vector Storage<\/h3> <p>In this tutorial, we will walk through building a Retrieval-Augmented Generation (RAG) system using the Jina.AI embeddings API and TiDB’s Vector Storage. This combination allows for efficient handling of vectorized data and powerful retrieval capabilities. We will cover the following steps: Prerequisites Before we begin, ensure you have the following: Step 1: Setting Up Your […]<\/p> <\/a>","_links":{"self":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/article\/17768","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/article"}],"about":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/types\/article"}],"author":[{"embeddable":true,"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/users\/8"}],"wp:attachment":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/media?parent=17768"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}} - Image Search<\/a>: use the OpenAI CLIP model to generate vectors for image and text.<\/li>\n\n\n\n
- OpenAI Embedding<\/a>: use the OpenAI embedding model to generate vectors for text data.<\/li>\n\n\n\n
- Follow this tutorial<\/a> to create a TiDB Serverless cluster with vector support enabled<\/li>\n\n\n\n
- Sign up TiDB Cloud<\/a><\/li>\n\n\n\n