the AI & cloud cost gateway

Don't let vibe coding eat your budget.

slice sits in front of every AI call your team makes, routes the easy ones to cheaper models, and cuts the bill before it runs away.

Early. One person, deploying now.
a slice of cake
click to cut
slice · measured run50-prompt dev workload
Easy edits, lint, one-linersrouted to Haiku$0.03
Refactors and unit testskept on Sonnet$0.11
Repeat promptsserved from cache$0.00
50 prompts, sent both ways, same models priced the same
Direct to Anthropic$0.26
Through slice$0.14
You saved
$0.12
45% less · same workload
SAVED

measured · the same 50 prompts, sent direct and through slice

get started

From clone to first request.

About three minutes. You sign in once with GitHub, then point your tools at slice.

(like this)  means a value you fill in with your own. Everything else you can copy as-is.

01

Install slice

slice is a small open-source package. Clone the repo and install it, and the slice command lands on your path.

Terminal
$ git clone https://github.com/jjk30/slice.git && cd slice && pip install -e .
02

Sign in with GitHub

Run slice login. slice shows you a short code. Open github.com/login/device, sign in to GitHub, paste the code, and click Authorize. That is the whole login, no password typed in the terminal.

Why GitHub? You already have it, slice never stores a password, and it signs you in right from the terminal, where slice lives.

Terminal
$ slice login

  slice needs to connect to your GitHub account.

  1. copy this code   WXYZ-1234
  2. open             https://github.com/login/device
  3. paste it, then click Authorize

  waiting for you to authorize...
   signed in as (your-github-username)
   your slice key is ready
03

Point your tools at slice

Set one environment variable so your AI tools send their calls through slice. From here slice routes, caches, and meters every request. No key to paste, your login already set that up.

Terminal
export ANTHROPIC_BASE_URL=https://api.sliceapp.dev
04

Use your tool as normal

That is it. Keep coding the way you already do. Every call now flows through slice, and you can watch spend, savings, and budgets live on your slice dashboard.

Open source. slice is yours to read line by line, or to run on your own box. It only reads the calls passing through it, never your codebase.

on every request

What slice runs.

Routeeach request to the cheapest model that works.
Cacherepeated calls, so you never pay twice for the same answer.
Capeach team's spend. Warn as it fills, block at the ceiling.
Recommenda cheaper model, or a flat-fee tool when one fits the work better.
Watchyour AWS bill and mistakes, read-only, right next to AI spend.
Public bucket exposed
s3://acme-user-uploads is open to the whole internet and holds files that look like IDs and passports. slice caught it before anyone else did.
before it's a problem

You'll hear it from us, not from finance.

Set a budget. The moment a team crosses the line, slice buzzes you on WhatsApp and email, with the number, the cause, and how long you have.

sliceWhatsApp
online
Budget warning. Your coding team just crossed 80% of this month's $6,000 cap. About $1,080 left. Reply STATUS for a breakdown.
9:42 AM ✓✓
Gmail
slice alertsalerts@sliceapp.dev
9:42 AM
Budget warning: coding team at 82%
You've spent $4,920 of $6,000 this month. At today's pace you'll hit the cap in about 5 days. Biggest driver: Opus on lint fixes, which slice can route down to Haiku…
Open dashboard →

Pick the channel, set the number, slice does the buzzing. Email and WhatsApp today, Slack next.

a note from me

Why I built slice.

The idea showed up while I was craving a slice of cake. The name stuck, and so did the problem.

All through 2026 I kept reading the same story: great AI tools, adopted fast, billed per token with no ceiling and nobody watching the meter. Uber torched its year's coding budget by April. Amazon burned $1.8M on a single task. That is not a tooling problem. It is a missing brake.

So I built the brake. slice meters every call, routes the easy ones to cheaper models, caches the repeats, and cuts you off before the invoice does.

It is early, and I would rather say so than pretend. One person, deploying now. Try it on your own machine and judge it for yourself.

the stories that got me obsessed

Try it. Decide for yourself.

Free and open source. Three commands from clone to your first metered request.

Free · open source