TENSORUS

World's First Tensor Database

The Agentic
Tensor Database

3
Engines
18
Properties
6D
Tensors
Get Started
Search Engines

Three specialized
approaches

HNSW

Vector Similarity

Hierarchical navigable small world graphs for fast approximate nearest neighbor search.

< 10ms
Query latency
ALEX

Property Search

Learned index structures optimized for filtering by mathematical properties.

18
Searchable properties
Tucker

Structural Similarity

Tensor decomposition for finding similar multi-dimensional structures.

Subspace dimensions
Mathematical Properties

Search by what
vectors can't

18 mathematical properties enable semantic search beyond simple vector similarity

6
Scalar Properties
Frobenius Norm
L1 Norm
L∞ Norm
+3 more
6
Matrix Properties
Rank
Trace
Determinant
+3 more
6
Boolean Properties
Symmetric
Positive Definite
Orthogonal
+3 more
Use Cases

Built for modern
data teams

AI/ML Engineers

Store and search embeddings with mathematical property filters for better model performance

Vector searchProperty filteringMulti-dimensional

Data Scientists

Analyze tensor structures and relationships across high-dimensional datasets

Tensor decompositionStructural similarityAnalytics

Research Teams

Query complex mathematical properties for scientific computing and simulations

18 propertiesFast queriesScalable

Platform Engineers

Build agentic systems that need intelligent tensor storage and retrieval

Agent-readyAPI-firstProduction-grade