Weaviate Has a Free API: Vector Database With Built-in ML Models for Semantic Search

Weaviate is a vector database that can vectorize your data automatically using built-in ML models. No need to manage embeddings separately — just send text and Weaviate handles the rest.

Why Weaviate?

• Auto-vectorization — built-in OpenAI, Cohere, HuggingFace integrations
• GraphQL + REST — query with GraphQL or REST API
• Hybrid search — combine BM25 keyword + vector similarity
• Generative search — RAG built into the database
• Free cloud — 14-day sandbox, self-host unlimited

Quick Start

# Docker
docker run -d -p 8080:8080 -p 50051:50051 \
  -e QUERY_DEFAULTS_LIMIT=20 \
  -e AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED=true \
  -e PERSISTENCE_DATA_PATH='/var/lib/weaviate' \
  -e CLUSTER_HOSTNAME='node1' \
  semitechnologies/weaviate:latest

# Python client
pip install weaviate-client

Python SDK

import weaviate
import weaviate.classes as wvc

client = weaviate.connect_to_local()  # localhost:8080

# Create collection with auto-vectorization
articles = client.collections.create(
    name="Article",
    vectorizer_config=wvc.config.Configure.Vectorizer.text2vec_openai(),
    generative_config=wvc.config.Configure.Generative.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),
    ],
)

# Insert — Weaviate auto-vectorizes the text!
articles.data.insert_many([
    {"title": "Intro to RAG", "content": "RAG combines retrieval with generation...", "category": "AI"},
    {"title": "Vector DBs", "content": "Vector databases store embeddings...", "category": "Database"},
])

# Semantic search (just send text, not vectors!)
results = articles.query.near_text(
    query="how to build AI search",
    limit=5,
    return_properties=["title", "content"],
)
for obj in results.objects:
    print(obj.properties["title"])

# Hybrid search (keyword + semantic)
results = articles.query.hybrid(
    query="vector database performance",
    alpha=0.5,  # 0=keyword, 1=vector
    limit=5,
)

# Generative search (RAG in one query!)
results = articles.generate.near_text(
    query="explain RAG",
    grouped_task="Summarize these articles in 2 sentences",
    limit=3,
)
print(results.generated)  # AI-generated summary using retrieved context

REST API

BASE="http://localhost:8080/v1"

# Schema
curl $BASE/schema

# Get objects
curl $BASE/objects?class=Article&limit=10

# Create object
curl -X POST $BASE/objects \
  -H 'Content-Type: application/json' \
  -d '{
    "class": "Article",
    "properties": {
      "title": "New Article",
      "content": "Article content here"
    }
  }'

GraphQL API

{
  Get {
    Article(
      nearText: { concepts: ["machine learning"] }
      limit: 5
    ) {
      title
      content
      _additional {
        distance
        certainty
      }
    }
  }
}

Key Features

Feature	Details
Vectorizers	OpenAI, Cohere, HuggingFace, Ollama
Search	Semantic, keyword, hybrid, generative
API	REST, GraphQL, gRPC
Filtering	Metadata + vector combined
Multi-tenancy	Built-in tenant isolation
Replication	Built-in for HA

Resources

• Weaviate Docs
• Python Client
• Weaviate Cloud

---

Building AI-powered search? Check my Apify actors or email [e-mail].