valeriansaliou/sonic

Analysis updated 2026-06-21

★ 21,205RustAudience · developerComplexity · 3/5Setup · moderate

Mindmap

mindmap
  root((sonic))
    What it does
      Text search
      Autocomplete
      Typo correction
    Tech Stack
      Rust
      Docker Cargo
      Sonic Channel protocol
    Use Cases
      Helpdesk search
      Per-user indexes
      CRM search
    Audience
      Backend developers
      Product teams
    Features
      80+ languages
      Microsecond queries
      Low RAM usage

mindmap root((sonic)) What it does Text search Autocomplete Typo correction Tech Stack Rust Docker Cargo Sonic Channel protocol Use Cases Helpdesk search Per-user indexes CRM search Audience Backend developers Product teams Features 80+ languages Microsecond queries Low RAM usage

Click or tap to explore — scroll the page freely

What do people build with it?

USE CASE 1

Add instant text search and autocomplete to a customer support helpdesk or live chat platform without running Elasticsearch.

USE CASE 2

Power per-user isolated search indexes in a SaaS CRM by using separate Sonic buckets per user.

USE CASE 3

Replace slow SQL LIKE queries in a product catalog with sub-millisecond Sonic lookups via the Sonic Channel protocol.

USE CASE 4

Build an autocomplete feature that handles typos and suggests completions as the user types, across 80+ languages.

What is it built with?

RustDockerCargo

How does it compare?

	valeriansaliou/sonic	cjpais/handy	nikivdev/flow
Stars	21,205	21,215	21,138
Language	Rust	Rust	Rust
Setup difficulty	moderate	moderate	easy
Complexity	3/5	3/5	2/5
Audience	developer	general	developer

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Requires running a Sonic server process and integrating via the Sonic Channel text protocol, no off-the-shelf REST API.

In plain English

Sonic is a small, fast search backend that you run alongside your application to power text search and auto-complete on your own data. It is pitched as a lightweight alternative to heavier search systems like Elasticsearch for cases where you do not need every advanced feature, just quick lookups across a lot of items. The README claims it can respond to queries in microseconds, uses roughly 30 megabytes of RAM under load, and was originally created in Nantes, France for use inside the Crisp customer-messaging platform, where it indexes around half a billion items on a modest server. A few ideas explain how it works. Sonic is described as schema-less and as an identifier index rather than a document index, meaning it does not store your actual records, only the words people might search for and the identifiers that point back to those records. Your application still keeps the real data in its own database, and Sonic just returns IDs that you then look up. Words are organized into collections and buckets, so for example you could have a separate bucket per user to keep their search indexes isolated. Items can be pushed into or popped from the index while the server is running, and a background process consolidates these changes so they appear in search shortly after. Sonic can suggest completions while a user types, fix small typos when there are not enough exact matches, and clean up stop words across more than eighty supported languages. You talk to the server using a simple text-based protocol called Sonic Channel, with client libraries available for common languages. It is a good fit for product teams who want fast search and suggestions inside a chat app, helpdesk, CRM, or similar product without operating a full search cluster. Sonic is written in Rust and can be installed from source, Cargo, Debian packages, or Docker images.

Copy-paste prompts

Prompt 1

I want to add search to my Node.js helpdesk app using Sonic. Walk me through running Sonic in Docker, pushing ticket titles to the index using the Sonic Channel protocol, and querying it from Node.js.

Prompt 2

Using Sonic, how do I keep each user's search index separate, and how do I push and pop items from a user's bucket while Sonic is running?

Prompt 3

I'm building an autocomplete feature with Sonic. Show me how to use the SUGGEST command over the Sonic Channel protocol and how to tune it for fast suggestions as the user types.

Prompt 4

What does Sonic's typo-correction feature do, how do I configure how aggressively it fixes small spelling mistakes, and what languages does it support?

Frequently asked questions

What is sonic?

A fast, lightweight search backend written in Rust that handles text search and autocomplete for your app in microseconds, without storing your actual records, just the searchable words and IDs pointing back to your data.

What language is sonic written in?

Mainly Rust. The stack also includes Rust, Docker, Cargo.

How hard is sonic to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is sonic for?

Mainly developer.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Scan in gitsafehub Deploy in gitdeployhub valeriansaliou on gitmyhub

Verify against the repo before relying on details.