Upsert data

This page shows you how to upsert documents into a collection. The REST request accepts an optional branch field. Omitting it writes to the default branch, main; a tag or alias is not a valid write target. Document field names are preserved as supplied in every SDK. The examples use dense_vector and sparse_vector; configure indexes with those same names to search these values.

If the collection contains managed embedding vector fields, send only the configured source text fields. Do not send direct vector values for managed embedding fields in an upsert request.

from lambdadb import LambdaDB

with LambdaDB(
    project_api_key="YOUR_API_KEY",
    base_url="YOUR_BASE_URL",
    project_name="YOUR_PROJECT_NAME",
) as client:
    coll = client.collection("my_collection")
    docs = [
        {"id": "33201222", "url": "https://en.wikipedia.org/wiki/LambdaDB", "title": "LambdaDB", "text": "LambdaDB is the vector lake for AI memory ... ", "dense_vector": [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0], "sparse_vector": {"LambdaDB": 0.83, "is": 0.1, "a": 0.1, "AI": 0.7}},
        {"url": "https://en.wikipedia.org/wiki/Winamp", "title": "Winamp", "text": "Winamp is a media player for Windows, macOS and Android ...", "dense_vector": [1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0], "sparse_vector": {"0": 0.8, "4": 0.1, "11": 0.1, "63": 0.54}},
    ]
    coll.docs.upsert(docs=docs)

Python: LambdaDB supports context manager usage. __enter__ returns the client, and __exit__ calls client.close() (closing the SDK-owned HTTP client) and makes the client unusable after the with block. If you don’t use with, call client.close() when you’re done. If you pass a custom client=/async_client=, you own closing it.

import { LambdaDBClient } from "@functional-systems/lambdadb";

const client = new LambdaDBClient({
  projectApiKey: "YOUR_API_KEY",
  baseUrl: "YOUR_BASE_URL",
  projectName: "YOUR_PROJECT_NAME",
});
const collection = client.collection("my_collection");
await collection.docs.upsert({
  docs: [
    { id: "33201222", url: "https://en.wikipedia.org/wiki/LambdaDB", title: "LambdaDB", text: "LambdaDB is the vector lake for AI memory ... ", dense_vector: [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0], sparse_vector: { LambdaDB: 0.83, is: 0.1, a: 0.1, AI: 0.7 } },
    { url: "https://en.wikipedia.org/wiki/Winamp", title: "Winamp", text: "Winamp is a media player for Windows, macOS and Android ...", dense_vector: [1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0], sparse_vector: { "0": 0.8, "4": 0.1, "11": 0.1, "63": 0.54 } },
  ],
});

package main

import (
  "context"
  "log"
  lambdadb "github.com/lambdadb/go-lambdadb"
)

func main() {
  ctx := context.Background()
  client := lambdadb.New(
    lambdadb.WithBaseURL("YOUR_BASE_URL"),
    lambdadb.WithProjectName("YOUR_PROJECT_NAME"),
    lambdadb.WithAPIKey("YOUR_API_KEY"),
  )
  coll := client.Collection("my_collection")
  myDocs := []map[string]interface{}{
    {"id": "33201222", "url": "https://en.wikipedia.org/wiki/LambdaDB", "title": "LambdaDB", "text": "LambdaDB is the vector lake for AI memory ... ", "dense_vector": []float64{0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0}, "sparse_vector": map[string]float64{"LambdaDB": 0.83, "is": 0.1, "a": 0.1, "AI": 0.7}},
    {"url": "https://en.wikipedia.org/wiki/Winamp", "title": "Winamp", "text": "Winamp is a media player for Windows, macOS and Android ...", "dense_vector": []float64{1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0}},
  }
  _, err := coll.Docs().Upsert(ctx, lambdadb.UpsertDocsInput{Docs: myDocs})
  if err != nil {
    log.Fatal(err)
  }
}

LAMBDADB_PROJECT_API_KEY="YOUR_API_KEY"

curl -X POST \
  "$BASE_URL/projects/$PROJECT_NAME/collections/{collectionName}/docs/upsert" \
  -H 'content-type: application/json' \
  -H "x-api-key: $LAMBDADB_PROJECT_API_KEY" \
  -d '{
    "branch": "candidate",
    "docs": [
      {
        "id": "33201222",
        "url": "https://en.wikipedia.org/wiki/LambdaDB",
        "title": "LambdaDB",
        "text": "LambdaDB is the vector lake for AI memory ... ",
        "dense_vector" : [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0],
        "sparse_vector" : {"LambdaDB": 0.83, "is": 0.1, "a": 0.1, "AI": 0.7}
      },
      {
        "url": "https://en.wikipedia.org/wiki/Winamp",
        "title": "Winamp",
        "text": "Winamp is a media player for Windows, macOS and Android ...",
        "dense_vector" : [1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0],
        "sparse_vector" : {"0": 0.8, "4": 0.1, "11": 0.1, "63": 0.54}
      }
    ]
  }'

The cURL example writes to the candidate branch. Remove branch to write to main.

Response

{
  "message": "Upsert request is accepted"
}

Each document implicitly contains an id field in order to uniquely identify a document. A unique string value is auto-generated by the system if id field is not provided in an upsert request. If you want to overwrite the entire document, you can do so by providing the id field in the document.

For a partitioned collection, documents are upserted into the __default__ partition if the partition field is not provided.

Upsert limits

Metric	Limit
Max payload size	6MB
Max length for a document ID	512 characters
Max vector dimensions	4,096
Max document size	5MB

When upserting larger amounts of data, it is recommended to use bulk-upsert operation.

Get started

Migrations

Collections

Data versioning

Documents

Search

Manage Costs

Response

Upsert limits

​Response

​Upsert limits

Response

Upsert limits