Frequently asked questions

The minimal commitment rate is the monthly base payment you commit to up front.

Your commitment level determines the discount applied to your usage rates — the higher the commitment, the higher the discount, up to 50%.

Each month, your billing total is whichever is greater: your discounted usage cost, or your commitment — plus any add-ons.

Standard support is always included, with next-business-day email support and access to our Discord channel.

From a monthly commitment of $5,000 onwards, 24/7 priority support via email and phone is included.

There are no rate limits for the SeekStorm library, the self-hosted SeekStorm server, or the SeekStorm SaaS running on dedicated infrastructure.

For the SeekStorm SaaS on shared infrastructure, API requests are rate limited to a maximum of 1,000 requests per second (2.6 billion requests per month), with a maximum of 20 concurrent requests.

When those rates are exceeded, our rate limiter may automatically block all requests from the affected API key, returning HTTP error code 429 (rate limit exceeded).

The block is lifted automatically once average traffic returns to 1 query/second. The more the limit was exceeded, the longer it takes for the average to return to that level — so the blocking duration scales with the intensity of the previous overload.

There are no limits for the SeekStorm library, the self-hosted SeekStorm server, or the SeekStorm SaaS on dedicated infrastructure, regarding the number of:

• indices
• documents
• fields
• field length
• terms per document

For the SeekStorm SaaS on shared infrastructure, the following limits apply:

• max indices per API key: 10
• max documents per API key: 100M
• max fields: 1,000
• max field length: 10M
• max terms per document: 1M

Additional limits apply to facets:

• maximum 65,535 (String16) or 4,294,967,295 (String32) distinct values per string facet field
• maximum 65,535 (StringSet16) or 4,294,967,295 (StringSet32) distinct value combinations per string set facet field
• maximum 65,536 distinct numerical ranges per facet field

Start small. Grow up. Never re-index. The same binary index moves with you — from an embedded library, to a self-hosted server, to fully managed cloud — and back.

In SeekStorm, an index is a self-contained set of files. You can copy, back up, or move it between environments — local, self-hosted, or cloud — without re-indexing and without vendor lock-in.

Transferring index data between a local/self-hosted setup and the SeekStorm SaaS is done via cloud storage services or a USB stick, and can be requested through our support form for a fee that depends on the index size.

The REST API is more low-level and requires you to use an HTTP client native to your stack. The requests and responses are encoded in JSON.

The SDKs are more high-level, where requests and responses are objects, idiomatic to your language. The HTTP client is already part of the SDK, without requiring any user attention. The SDKs are available in multiple languages (Python, TypeScript, Rust, C#, Java, Go).

The general functionality of the API endpoints stays the same, regardless of whether accessed via SDK or REST.

Index backup is optional and can be enabled in the cost calculator. Its cost depends on the number of indexed documents and the average size per document.

Restoring from a backup snapshot protects against data loss and lets you skip the expensive, time-consuming process of rebuilding the index from scratch.

SeekStorm uses a multi-stage retention scheme (the GFS — Grandfather-Father-Son — concept):

A backup snapshot contains the entire index directory, including the document store, the lexical index, and the vector index. SeekStorm backups are full backups, not incremental.

Restoring an index from a backup can be requested via the support form.

Your index can be hosted on dedicated infrastructure or on shared infrastructure. Shared infrastructure is implemented via multi-tenancy: multiple users (tenants) are served by a single instance, sharing the same underlying hardware.

Dedicated infrastructure

• Isolated infrastructure: your data and operations run entirely on a dedicated instance, so no "noisy neighbor" effects from other tenants can slow you down.
• Enhanced security & compliance: stronger physical and logical data isolation, since your data stays completely separate from other customers'.
• Higher reliability and availability
• Predictable performance
• Higher capacity for both index size and throughput

Shared infrastructure

• More cost-efficient, since resources are pooled across multiple users. Not everyone needs all resources at the same time, so multi-tenancy makes use of capacity that would otherwise sit idle.
• Faster to spin up and wind down, making it well suited to workloads that scale quickly or vary a lot over time.

SeekStorm offers both options.