Weaviate Has a Free API — Here's How to Build AI-Powered Search with Vector Databases

Why Weaviate?

Weaviate is an open-source vector database for AI applications. It stores data as vectors (embeddings) and enables semantic search — find results by meaning, not just keywords.

Weaviate Cloud free tier (sandbox): 500K objects, 14-day persistence.

Getting Started

Option 1: Weaviate Cloud (Free Sandbox)

Sign up at weaviate.io — get a free sandbox cluster.

Option 2: Docker

docker run -d -p 8080:8080 -p 50051:50051 \
  -e QUERY_DEFAULTS_LIMIT=25 \
  -e AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED=true \
  -e DEFAULT_VECTORIZER_MODULE=text2vec-transformers \
  -e ENABLE_MODULES=text2vec-transformers \
  semitechnologies/weaviate:latest

Python Client

import weaviate
import weaviate.classes as wvc

client = weaviate.connect_to_weaviate_cloud(
    cluster_url="https://your-sandbox.weaviate.network",
    auth_credentials=weaviate.auth.AuthApiKey("YOUR_API_KEY")
)

# Create collection with auto-vectorization
client.collections.create(
    name="Article",
    vectorizer_config=wvc.config.Configure.Vectorizer.text2vec_openai(),
    properties=[
        wvc.config.Property(name="title", data_type=wvc.config.DataType.TEXT),
        wvc.config.Property(name="content", data_type=wvc.config.DataType.TEXT),
        wvc.config.Property(name="category", data_type=wvc.config.DataType.TEXT)
    ]
)

articles = client.collections.get("Article")

# Add data (auto-vectorized!)
articles.data.insert_many([
    {"title": "Introduction to Machine Learning", "content": "ML is a subset of AI that learns from data...", "category": "AI"},
    {"title": "Web Scraping Best Practices", "content": "Learn how to scrape websites ethically...", "category": "Dev"},
    {"title": "Building REST APIs with FastAPI", "content": "FastAPI is a modern Python framework...", "category": "Dev"},
    {"title": "Neural Networks Explained", "content": "Deep learning uses layers of neurons...", "category": "AI"}
])

# Semantic search — finds by MEANING, not keywords!
results = articles.query.near_text(
    query="artificial intelligence and deep learning",
    limit=3
)

for obj in results.objects:
    print(f"{obj.properties['title']} [{obj.properties['category']}]")
# Returns ML and Neural Networks articles — even without exact keyword match!

# Hybrid search (vector + keyword)
results = articles.query.hybrid(
    query="Python web framework",
    limit=3,
    alpha=0.5  # 0=keyword, 1=vector, 0.5=balanced
)

# Filtered vector search
results = articles.query.near_text(
    query="learning algorithms",
    filters=wvc.query.Filter.by_property("category").equal("AI"),
    limit=5
)

JavaScript Client

import weaviate from "weaviate-client";

const client = await weaviate.connectToWeaviateCloud(
  "https://your-sandbox.weaviate.network",
  { authCredentials: new weaviate.ApiKey("YOUR_KEY") }
);

const articles = client.collections.get("Article");

// Semantic search
const result = await articles.query.nearText("machine learning basics", { limit: 5 });
result.objects.forEach(obj => console.log(obj.properties.title));

// Generate (RAG — Retrieval Augmented Generation)
const ragResult = await articles.generate.nearText(
  "AI concepts",
  { singlePrompt: "Summarize this article in one sentence: {title} - {content}" },
  { limit: 3 }
);

ragResult.objects.forEach(obj => {
  console.log(`${obj.properties.title}`);
  console.log(`Summary: ${obj.generated}`);
});

Use Cases

1. Semantic search — find products/docs by meaning
2. RAG — feed context to LLMs from your data
3. Recommendation — "similar items" based on embeddings
4. Image search — search images by description
5. Anomaly detection — find outliers in vector space

---

Need to scrape data for your vector database? I build production-ready scrapers. Check out my Apify actors or email [e-mail] for custom data pipelines.

Building AI search? Share your vector DB experience below!