HyperspaceDB icon

HyperspaceDB

Interact with the HyperspaceDB gRPC API

Collection → Create

AI-generated

Summary

Create a new collection in HyperspaceDB by specifying its name, vector dimension, and distance metric.

Inputs

  • Collection Name (required) — The unique name to assign to the new collection.
  • Dimension (required) — The number of vector dimensions the collection will store (e.g., 1024 for Hyperspace default).
  • Metric (required) — The distance metric used for similarity calculations; options are Cosine, L2 (Euclidean), Poincare, or Lorentz.

Output shape

a single record indicating creation success and the collection name

Returns an object like { success: true, name } confirming the collection was created; throws error if creation fails.

Examples

Example 1: Create a collection named 'myCollection' with 1024 dimensions and cosine similarity.

Set 'Collection Name' to 'myCollection', 'Dimension' to 1024, and 'Metric' to 'cosine'.

Links

Discussion