95% fewer answer input tokensRead
Serverless search at any scale

Search your data.
Version every change.

Vector, full-text, and hybrid search with built-in data versioning. Branch, test, and pin data versions for AI applications.

Set up with your AI agent

Copy this prompt into your coding agent.

Set up LambdaDB for this project. Read https://docs.lambdadb.ai/guides/get-started/with-an-agent.md and follow the setup instructions.

Read setup guide
Get API key Read the Docs
A conceptual view of LambdaDB: documents, versions, vectors, and metadata in searchable storage, with continuous ingest and on-demand search.documentsversionsvectorsmetadatadocs · vectors · metadata · versionsSEARCHABLE DATAcontinuous ingestcontinuous ingeston-demand searchon-demand searchnothing running when idle
Search latency
280 ms
P99 @ 128 QPS · 22M vectors
Ingest speed
19 s
to ingest 1M vectors
Compute cost
$0
when idle
Version control

Test your data like you test your code.

Data changes. Test those changes before they reach your application.

Branch your data to experiment, pin versions for evaluation, and switch reads when you are ready.

What one version history unlocks

Swap embedding models on a branch, and promote only what beats production.

Test updates to agent memory on a branch before switching reads.

Pin reads to a tag to evaluate searches against the same data version.

Four steps on one collection timeline, snapshot-adoption model. Step 1: tag v1 pins the current state and production serves it. Step 2: a branch forks off copy-on-write. Step 3: evals pass. Step 4: the branch snapshot is adopted wholesale as version v2 of the collection, with no merge and no conflicts; the branch is deleted and the production alias moves from v1 to v2.alias: productionalias: productionkb1tag v1tag v2v1 stays for rollback2embed-v2 · copy-on-write3evals passed4adopt the snapshot as v2whole state · no merge · no conflicts
The same four steps, laid out vertically for small screens. Step 1: tag v1 pins the current state and the production alias hangs under it. Step 2: a branch forks off copy-on-write. Step 3: evals pass on the branch. Step 4: the branch snapshot is adopted as tag v2 and the production alias now hangs under v2.kbtag v11production2embed-v2 · copy-on-writeevals passed34tag v2production
1Tag v1 pins today's state. Production serves it.2Branch off copy-on-write: full data, no copy.3Run evals on the branch until they pass.4Adopt the branch snapshot as v2 (no merge). Delete the branch; production moves to v2.
The primitives underneath
Tags

Freeze a state as an immutable version.

Aliases

Point traffic at a version and repoint without code changes.

Branches

A connected line of versions inside a collection.

Versioned Search

Search any version: branch, tag, or alias.

Reproducible Reads

Pin reads to a tag to search the same data version again.

Scaling

Scales with your workload. $0 compute when idle.

Hybrid search fans out one request into parallel queries; agents stack writes on top of every read. The burst lands on your biggest day.

Provisioned capacity can leave you paying for idle resources, while request quotas can constrain a burst. LambdaDB manages capacity for you.

A conceptual scaling illustration, not measured performance data: workload (solid ink), capacity and quota limits (dashed), and on-demand capacity (orange). The shaded regions illustrate idle capacity, degraded service, and requests above a quota. Actual service limits apply.your workloadvendor limitsLambdaDB capacitydb traffic ↑time →idle · still billedserverless vendor quotaprovisioned capacityoutagedegradedvendor quotaprovisioned capacityLambdaDBidle · still billedLambdaDB capacity
idle you still pay fordegraded: above provisioned capacityoutage: above vendor quotaLambdaDB: scales with demand
Scaling illustration. View service limits →
Features

Production-grade search, included in every plan.

No platform fees or minimum monthly charges. Hybrid search, filters, consistent reads, and language-specific text analyzers are available on every plan, including free.

From text and images to relevant results

Generate embeddings from text and images, combine full-text and vector search with filters, and rerank results with native integrations. Bring your own vectors if you prefer.

Native embeddings Native reranking
ts
// Hybrid search with a native embedding field
const results = await client.collection("kb").query({
  size: 10,
  query: {
    rrf: [
      { queryString: {
          query: 'title:(refund OR policy)^2.5 OR content:lambdadb~2' } },
      { knn: {
          filter: { queryString: { query: "metadata.type:guide" } },
          field: "content_embedding",
          queryText: "refund policy update",
          k: 10 } },
    ],
  },
});
// Hybrid search with native embeddings
const kb = client.collection("kb");
const results = await kb.query({
  size: 10,
  query: {
    rrf: [
      { queryString: {
          query:
            'title:policy^2 OR refund~1' } },
      { knn: {
          filter: { queryString: {
            query: "metadata.type:guide" } },
          field: "content_embedding",
          queryText: "refund policy",
          k: 10 } },
    ],
  },
});

Works with your AI stack. MCP server for Claude, Cursor, and any MCP client · Python, TypeScript, and Go SDKs · REST · LangChain · MCP setup guide →

Guides for migrating from Elasticsearch, Pinecone, and Qdrant →

Safety

Isolated, encrypted, recoverable.

What happens to your data, in plain terms.

Isolation

API keys are scoped to one project. A key can't read another project's data.

Data residency

Data lives in the AWS region you choose, and stays there.

Durability

Object storage underneath, replicated across availability zones.

Recovery

Roll back to any tagged version, and search past versions to see what changed.

Encrypted at rest and in transit · SOC 2 Type II in progress.

Stay on the Frontier

Oct 25, 2026

AI Memory with 95% Fewer Tokens, Without Graphs or LLM Bookkeeping

Read more →
Oct 6, 2026

Native Reranking with Jev for More Relevant Results

Read more →

FAQFrequently asked questions

LambdaDB is a serverless search database combining vector, full-text, and hybrid search with built-in data versioning.

You pay for storage, reads, and writes — compute is $0 between requests. Your documents, vectors, and indexes live in object storage, a durable layer with no compute attached. Query and indexing workers spin up per request and disappear when the burst passes, so there are no always-on index servers to bill.

Create a branch to isolate changes, use a tag to pin a version, and point an alias at the branch or tag your application should read. You can search a selected version and switch the alias when you're ready.

Explore data versioning →

It is, if you hit cold storage on every query. LambdaDB keeps frequently-read data in a cache layer the system manages for you, so most reads never touch the object store. Measured on 22M Cohere-V3 vectors (1024-dim): P99 280 ms at 128 QPS.

Limits and pricing in the docs →

In durable object storage, in the region you choose, isolated per project. A write is durable once acknowledged and becomes searchable within seconds. When a query needs to see its own writes immediately, enable consistent reads — conditions and limits are in the search guide.

Search guide →

Either way works. Bring your own vectors, or use native embeddings to generate vectors from text and images and embed queries with the configured model.

LambdaDB combines vector, full-text, and hybrid search with data versioning. Branch data to test changes, pin versions with tags, and switch application reads with aliases. Its serverless architecture uses object storage and on-demand compute, with no always-on compute charge when idle.

Didn't find your question? Ask in the community Slack or talk to us.

Start small. Experiment safely.
Scale what works.

Hybrid search with no clusters to size. Branch, evaluate, and promote with aliases.

Get API key Read the Docs

View pricing →