Skip to main content
Adds new vectors to the index or updates existing ones. Supports both dictionary format and separate arrays for IDs and vectors. 
# Option 1: Dictionary format
index.upsert(items)

# Option 2: Separate arrays  
index.upsert(ids, vectors)

​
Parameters

​
Option 1: Dictionary Format

ParameterTypeDefaultDescription
itemsList[Dict]-List of dictionaries containing vector data
Where each dictionary can contain: 
[
  {
    "id": str, # Unique identifier for the vector
    "vector": List[float], # The vector data
    "contents": str | bytes, # Optional content (accepts both strings and bytes). Bytes are base64-encoded, strings are passed directly. All contents are encrypted before storage.
    "metadata": Dict # Optional key-value pairs for filtering and retrieval
  },
  ...
]
The contents field accepts both strings and bytes. Bytes are automatically base64-encoded before encryption, while strings are passed as-is. Contents are returned in their original format (string or bytes) when retrieved with get().

​
Option 2: Separate Arrays

ParameterTypeDefaultDescription
idsList[str]-List of unique vector identifiers
vectorsList[List[float]] or np.ndarray-Vector data as 2D array
metadataList[Dict]None(Optional) List of metadata dictionaries, one per vector
contentsList[str | bytes]None(Optional) List of content values, one per vector. Bytes are base64-encoded, strings are passed directly. All contents are encrypted before storage
When np.ndarray is passed as vectors, they are automatically sent in an optimized binary format for better performance with large batches.

​
Returns

None

​
Example Usage

​
Dictionary Format

# Basic vector upsert
items = [
    {
        'id': 'doc1',
        'vector': [0.1, 0.2, 0.3, 0.4]
    },
    {
        'id': 'doc2', 
        'vector': [0.5, 0.6, 0.7, 0.8]
    }
]

index.upsert(items)

​
Separate Arrays

import numpy as np

# Using separate arrays
ids = ['vec1', 'vec2', 'vec3']
vectors = np.random.rand(3, 128).astype(np.float32)

index.upsert(ids, vectors)