Now in Public Beta

The Knowledge Layer for AI

Hypergraph database for AI/ML that enables multi-hop reasoning across complex relationships. Store facts, entities, and episodes with native vector search.

terminal
# Install HyperX
$ pip install hyperx
# Connect to HyperX and add knowledge
from
hyperx import HyperX

db = HyperX(api_key="hx_sk_...")

# Create entities
alice = db.entities.create(name="Alice", entity_type="person")
bob = db.entities.create(name="Bob", entity_type="person")
project = db.entities.create(name="Project-X", entity_type="project")

# Create a hyperedge connecting multiple entities
edge = db.hyperedges.create(
description="collaborates on",
members=[
{"entity_id": alice.id, "role": "collaborator"},
{"entity_id": bob.id, "role": "collaborator"},
{"entity_id": project.id, "role": "project"},
]
)

# Multi-hop query
paths = db.paths.find(start=alice.id, hops=2)

Built for AI Applications

Everything you need to build intelligent systems that understand context and relationships.

Hyperedges

Connect any number of nodes in a single edge. Model complex N-ary relationships that traditional graphs cannot express.

Multi-hop Paths

Traverse relationships across multiple hops to discover hidden connections. Enable reasoning chains for RAG pipelines.

Hybrid Search

Combine vector similarity with graph traversal. Find semantically related knowledge within relationship context.

Memory Types

Built-in support for facts, entities, episodes, procedures, preferences, and goals. Structure knowledge the way humans think.

Bi-Temporal

Track when facts were true and when they were recorded. Query knowledge as it existed at any point in time.

Native Embeddings

Store and query vector embeddings directly on nodes and edges. Powered by efficient cosine similarity search.

Pay for What You Use

Simple, usage-based pricing. Start free, scale as you grow. No upfront costs.

Free Tier

Get started at no cost

$0 /month
10K
Entities
50K
API calls/month
100MB
Vector storage
Start Free

Ready to Build Smarter AI?

Join developers building the next generation of AI applications with HyperX.