Skip to content
HeliosDB HeliosDB Full

Pinecone Vector Protocol Documentation

This directory contains consolidated documentation for HeliosDB’s Pinecone vector database protocol support.

Quick Start

Connect to HeliosDB using the Pinecone client:

from pinecone import Pinecone


# Connect to HeliosDB (Pinecone-compatible)
pc = Pinecone(api_key="your-api-key", host="http://localhost:8080")


# Access index
index = pc.Index("my-vectors")


# Upsert vectors
index.upsert(vectors=[
    {"id": "vec1", "values": [0.1, 0.2, 0.3, ...], "metadata": {"category": "A"}},
    {"id": "vec2", "values": [0.4, 0.5, 0.6, ...], "metadata": {"category": "B"}}
])


# Query similar vectors
results = index.query(vector=[0.1, 0.2, 0.3, ...], top_k=10)

Contents

FileDescription
README.mdOverview and quick start (this file)
CONFIGURATION.mdConnection and API configuration
COMPATIBILITY.mdPinecone API compatibility
EXAMPLES.mdVector search examples

Feature Overview

API Compatibility

OperationResultNotes
UpsertSupportedBatch operations
QuerySupportedTop-K search
FetchSupportedBy ID
DeleteSupportedBy ID or filter
UpdateSupportedMetadata update
ListSupportedPagination
Describe IndexSupportedIndex stats

Vector Search Features

  • Similarity Metrics: Cosine, Euclidean, Dot product
  • Filtering: Metadata-based filtering
  • Namespaces: Logical partitioning
  • Sparse Vectors: Hybrid search support
  • Batching: Bulk operations

Connection Parameters

ParameterDefaultDescription
hostlocalhostServer hostname
port8080Vector API port
api_key-API authentication
index-Index name

Use Cases

  • Semantic Search: Natural language queries
  • Recommendations: Similar item lookup
  • Image Search: Visual similarity
  • RAG Applications: Retrieval-augmented generation
  • Anomaly Detection: Outlier identification