Cargo
Leaderboard

A heavyweight Rust competition

That’s a lot
of target/.

Rust builds take up space. Compare your build artifacts, biggest cleans, and longest waits.

What are we measuring?

Total artifact footprint after a successful build.

Loading…
Biggest Rust builds, ranked by total artifact bytes
Rank Project / builder Build footprint Build time Recorded

Weighing the competition.

Fetching the latest community entries.

Personal best per project. All time.GitHub accounts. Self-reported scores.

Put your project
on the scales.

Your agent installs the CLI, connects GitHub, and submits your first build. Start with the prompt.

Your GitHub account, project label, measurements, platform and Cargo version are public. Source code and local paths stay on your machine.

Read the CLI guide
Agent setup
Paste into your coding agent

Open your Rust project in Codex, Claude Code, or another coding agent with terminal access. 

Set up Cargo Leaderboard in this Rust project using https://cargo-leaderboard.vercel.app/agents.md, then run a build and verify the submission.
Read the agent instructions
Prefer a terminal? Manual setup

Prebuilt binary, checksum verified. Requires Rust to build your projects.

1. Install 
curl -fsSL https://cargo-leaderboard.vercel.app/install.sh -o /tmp/cargo-leaderboard-install.sh &&
sh /tmp/cargo-leaderboard-install.sh
2. Connect your GitHub account 
cargo leaderboard login
3. Run inside your Rust project 
cargo leaderboard build
cargo leaderboard clean

Clean deletes your build artifacts, just like cargo clean. Setup trouble? Run cargo leaderboard doctor.

How the scores work.

Builds take up space.

We count file sizes across Cargo’s target and build directories after a successful build. Cached artifacts, dependencies and other profiles count too. This is total footprint, not just the latest build’s output.

Cleans give it back.

The before-and-after difference is your clean score. Partial cleans count only the decrease. Empty cleans don’t rank. Values use binary units: 1 GiB is 1,073,741,824 bytes.

A competition. Not a benchmark.

Hardware, flags and warm caches vary. Sizes are logical bytes, not allocated disk blocks. Shared directories include other projects. Avoid concurrent builds while measuring. GitHub accounts are verified; scores are self-reported.

Keeping a project private? Use --repo my-project to replace its label, or --no-submit to measure locally. Put wrapper options before Cargo flags.