// $ npm install turbopuffer import { Turbopuffer } from "@turbopuffer/turbopuffer" // API tokens are created in the dashboard https://turbopuffer.com/dashboard const TURBOPUFFER_API_KEY = process.env.TURBOPUFFER_API_KEY const tpuf = new Turbopuffer({ apiKey: TURBOPUFFER_API_KEY as string, // Pick the right region https://turbopuffer.com/docs/regions baseUrl: "https://gcp-us-central1.turbopuffer.com", }); const ns = tpuf.namespace(`vector-ts-1-${Date.now()}`); await ns.upsert({ vectors: [{ id: 1, vector: await openaiOrRandVector("A cat sleeping on a windowsill"), attributes: { "text": "A cat sleeping on a windowsill", "category": "animal" } }, { id: 2, vector: await openaiOrRandVector("A playful kitten chasing a toy"), attributes: { "text": "A playful kitten chasing a toy", "category": "animal" } }, { id: 3, vector: await openaiOrRandVector("An airplane flying through clouds"), attributes: { "text": "An airplane flying through clouds", "category": "vehicle" } }], distance_metric: "cosine_distance" }); // Basic vector search let results = await ns.query({ vector: await openaiOrRandVector("cat-like animal"), top_k: 2, distance_metric: "cosine_distance", include_attributes: ["text"] }); // Returns cat and kitten documents, sorted by vector similarity console.log(results); const ns2 = tpuf.namespace(`vector-ts-2-${Date.now()}`); // Advanced Example await ns2.upsert({ vectors: [{ id: 1, vector: await openaiOrRandVector("A shiny red sports car"), attributes: { "description": "A shiny red sports car", "color": "red", "type": "car", "price": 50000 } }, { id: 2, vector: await openaiOrRandVector("A sleek blue sedan"), attributes: { "description": "A sleek blue sedan", "color": "blue", "type": "car", "price": 35000 } }, { id: 3, vector: await openaiOrRandVector("A large red delivery truck"), attributes: { "description": "A large red delivery truck", "color": "red", "type": "truck", "price": 80000 } }, { id: 4, vector: await openaiOrRandVector("A blue pickup truck"), attributes: { "description": "A blue pickup truck", "color": "blue", "type": "truck", "price": 45000 } }], distance_metric: "cosine_distance" }); // Advanced vector search with filters results = await ns2.query({ vector: await openaiOrRandVector("car"), // Embedding similar to "car" top_k: 10, distance_metric: "cosine_distance", // Complex filter combining multiple conditions filters: ["And", [ ["color", "Eq", "blue"], ["price", "Lt", 40000], ["type", "Eq", "car"] ]], include_attributes: ["description", "price"] }); // Returns only blue cars under $40k, sorted by similarity to the query vector console.log(results); // Vector search with multiple vectors results = await ns2.query({ // Search for vehicles similar to either a car or truck vectors: [ await openaiOrRandVector("car"), await openaiOrRandVector("truck") ], top_k: 3, distance_metric: "cosine_distance", vector_query_mode: "or", // Match any of the query vectors include_attributes: ["description"] }); console.log(results); async function openaiOrRandVector(text: string): Promise<number[]> { if (!process.env.OPENAI_API_KEY) { console.log("OPENAI_API_KEY not set, using random vectors"); return [Math.random(), Math.random()]; } try { const { OpenAI } = await import('openai'); return (await new OpenAI().embeddings.create({ model: "text-embedding-3-small", input: text })).data[0].embedding; } catch { console.log("OpenAI package not installed, using random vectors (`npm install openai`)"); return [Math.random(), Math.random()]; } }

Navigation

Vector Search Guide