WarpSearch

npx add-mcp 'https://api.warpsearch.io/mcp'

LIVEPRODUCTION LATENCY --:-- UTC

Average query – ms

p99 – ms

Average query time per 30 minutes, last 24 hours. Updates every 15 minutes.

Built for what you’re building

All use cases

OpenSearch Serverless alternative Same OpenSearch clients and queries, no OCUs. From $10 a month instead of about $350, with millisecond search. Elasticsearch Serverless alternative Serverless Elasticsearch that bills per request, with no VCU baseline. Your clients, Logstash and Beats keep working. RAG retrieval Keyword and vector retrieval in one query against one index, so there is no separate vector database to keep in sync. E-commerce and site search Index each product with all its variants, filter on size, color and stock, and still match what shoppers mean. Connectors for Shopify, WooCommerce, Magento and more. Enterprise knowledge search Search wikis, tickets, docs and email in one place. Hybrid matching finds the answer even when people don't use your terms. Vector database alternative Filtered vector search with keyword and hybrid built in, and a turbopuffer-compatible API. From $10 a month. Log analytics Keep months of logs searchable, ship them with the tools you already run, and query them with Elasticsearch DSL, SQL or LogQL. SIEM and threat hunting Hunt across a year of security events. Firewall, DNS, IDS and endpoint logs in one place, at storage prices. Agent memory Agents create a namespace per user over MCP, and idle ones cost only their storage, so keeping one per user stays cheap.

Search it yourself.

Millions of documents. Your next keystroke.

Loading the live datasets...

Open the full playground

Full API compatibility with Elasticsearch, OpenSearch and Solr.

Works with your existing Elasticsearch, OpenSearch and Solr clients, any version. Change the endpoint, keep your code and pay fraction of the price. Moving off a serverless plan? See the OpenSearch Serverless alternative and the Elasticsearch Serverless alternative.

# your existing Elasticsearch query, sent to WarpSearch
curl -X POST "[YOUR WARPSEARCH ENDPOINT]/products/_search" \
  -H "Authorization: Bearer ws_live_…" \
  -H "Content-Type: application/json" \
  -d '{
    "query": { "match": { "title": "wireless headphones" } },
    "size": 3
  }'

# response
{
  "took": 5,
  "hits": {
    "total": { "value": 128, "relation": "eq" },
    "hits": [ … ]
  }
}

Point your agent at our MCP server. It handles the rest.

Your coding agent signs you up, creates a collection, loads your documents and runs the searches. You approve one email.

1. Add WarpSearch in your terminal

claude mcp add --transport http warpsearch https://api.warpsearch.io/mcp

2. Ask your agent

› set up search for our help-center articles
  register                        approved by email
  create_collection help_center   ok
  index_documents 248,316 docs    ok
  search "reset password"         5 ms

1. Add WarpSearch in your terminal

codex mcp add warpsearch --url https://api.warpsearch.io/mcp

2. Ask your agent

› set up search for our help-center articles
  register                        approved by email
  create_collection help_center   ok
  index_documents 248,316 docs    ok
  search "reset password"         5 ms

1. Add WarpSearch in your terminal

gemini mcp add --transport http warpsearch https://api.warpsearch.io/mcp

2. Ask your agent

› set up search for our help-center articles
  register                        approved by email
  create_collection help_center   ok
  index_documents 248,316 docs    ok
  search "reset password"         5 ms

1. Add WarpSearch to.vscode/mcp.json

{
  "servers": {
    "warpsearch": { "type": "http", "url": "https://api.warpsearch.io/mcp" }
  }
}

2. Ask your agent

› set up search for our help-center articles
  register                        approved by email
  create_collection help_center   ok
  index_documents 248,316 docs    ok
  search "reset password"         5 ms

Pay only for what you use.

Fully serverless search with nothing to run or maintain. Start free, grow without resizing anything, and never pay for idle capacity.

Fully serverless

No nodes, shards or replicas to size. Create an index and query it.

Zero maintenance

No upgrades, patching or rebalancing. We run the infrastructure, you ship search.

Scales to zero

An idle index costs only its storage. Traffic spikes are handled for you.

Usage-based billing

Billed on what you store, write and search, with volume discounts applied automatically. See pricing

Runs on AWS, Google Cloud and Azure.

Pick the provider and region closest to your app and your users. Each cloud is available in multiple regions, so your queries don't cross an ocean to be answered.

Bring your own cloud.

Need your data to stay in your own account? WarpSearch can also run inside your AWS, Google Cloud or Azure account, within your network and compliance boundary, with the same APIs.