LibreAI / Easy Install / llm_cost_tracker
// easy-install · sergey-homenko/llm_cost_tracker

llm_cost_tracker

Track LLM spend per user and feature inside your Rails app — no proxy.

💎 Hidden gem★ 44 stars⑂ 4 forksMITupdated 2026-10
▦ visual — this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
// project facts

Project facts

llm_cost_tracker is an open-source easy install project by sergey-homenko (sergey-homenko/llm_cost_tracker on GitHub). It has 44 GitHub stars and 4 forks, is released under the MIT license, and was last updated on GitHub on 4 October 2026.

GitHub stars
44
Forks
4
License
MIT
Last updated on GitHub
4 October 2026
Category
Easy Install
Listing
Hidden gem — a shorter page for a lesser-known project
Official site
Official site
// how it works

How it works

  1. Add the gem

    Include it in your Gemfile and bundle install.

  2. Run the migrations

    Create the cost-tracking tables in your database.

  3. Tag your calls

    Label LLM usage by user, feature, or custom tags.

  4. Review spend

    Check per-tag costs and set budgets.

// features

Why people love it

Per-user tracking

Attribute LLM spend to individual users.

Per-feature budgets

Cap costs by feature or any custom tag.

Your own database

Cost data stays in your Postgres, no third party.

No proxy required

Tracks calls without rerouting traffic.

Multi-provider

Anthropic, DeepSeek, Gemini, and more.

Rails native

Installs as a gem with migrations.

Advertisement
// run it

Run it yourself

  1. Step 1 — Add the gem

    Include it in your Gemfile.

    gem 'llm_cost_tracker'
  2. Step 2 — Install and migrate

    Bundle it and create the tracking tables.

    bundle install
    rails db:migrate
  3. Step 3 — Tag a call

    Label LLM usage in your code per the README.

    # see README for the tracking API
// use cases

What you can build

Advertisement
// faq

Questions, answered

Who maintains llm_cost_tracker?

It is an open-source project at sergey-homenko/llm_cost_tracker with 4 forks.

Is it free for commercial use?

Yes. It is MIT licensed.

Does it proxy my API calls?

No. It tracks spend without sitting in front of your model calls.

Which providers are supported?

Anthropic, DeepSeek, Gemini, and others via Faraday — see the README.

Where is the data stored?

In your own Rails database, via the gem’s migrations.

// keep exploring

Similar tools

Local LLM

agent-console

Local-first dashboard of tokens, models, and cost for coding agents.

★ 800 starsMIT
View guide →
AI Agents

superlog

Observability that dispatches AI agents to self-heal your software.

★ 1.5k starsApache-2.0
View guide →
Self-Hosted

analytics

Open source, privacy-first web analytics. Lightweight, cookie-free Google Analytics alternative. Self-hosted or cloud.

★ 29.3k starsAGPL-3.0
View guide →
Easy Install

Routerly

Self-hosted LLM gateway routing across providers with cost controls.

★ 99 starsAGPL-3.0
View guide →
Easy Install

gymcoach

A self-hosted AI workout tracker with your own LLM.

★ 31 starsMIT
View guide →
Easy Install

bfree-trainer-suite

Bno-cost Smart Trainer App 2026.

★ 56 starsCustom
View guide →
// alternatives

Alternatives in Easy Install

The most-starred other Easy Install projects in the LibreAI catalogue — the realistic shortlist if llm_cost_tracker is not the right fit.

Get llm_cost_tracker

Free and open — grab the source, star the repo, and start creating.