agentperf

How fast is your computer for AI coding agents?
agentperf benchmarks the local work behind every tool call,
then tells you what is slowing it down.

Quick start · What you get · Reading your score · Commands · How it works

When Claude Code or Codex works on your code, the model does its thinking in the cloud. Everything else happens on your computer. Each tool call starts a shell, reads or edits files, searches with ripgrep, runs git, and pipes the output back. One task can make hundreds of these calls, so a few milliseconds per call add up quickly.

agentperf times that local work. It runs the same fixed tests on every machine, compares your results with a reference machine, and explains your score: which tests cost you points, what is probably causing it, and what to fix first.

It is one Python package with no dependencies. It runs on macOS, Linux, WSL2, and Windows with Python 3.8 or newer.

Quick start

pipx install git+https://github.com/SergeSerb2/agentperf.git
agentperf run --quick --html report.html

A quick run takes a few minutes. When it finishes, open report.html and read Biggest wins first.

If you prefer not to install anything, download agentperf.pyz from the latest release and run it with Python:

python3 agentperf.pyz run --quick --html report.html

On Windows, type py instead of python3. In WSL, run agentperf inside the Linux distro, from a folder on the Linux filesystem (not under /mnt/c).

What you get

Every run gives you three numbers:

After the numbers, agentperf explains the score. This is the end of a run on a busy laptop:

AgentMark 430  (90% within-run interval 426-432, coverage 100%)  vs reference 'WSL2 - i7-12700K (reference, 1000)'
Reach     2,144  (network to model APIs and CDNs)
LW100     2.6 s local wait per 100 tool calls (reference 0.6 s)

Why this score
  Slow host. Agents spend noticeable time waiting on local tools; see the biggest wins.
  Biggest drag: process.parallel at x0.13 vs reference (worth up to +52 AgentMark).
  Biggest wins (gains are upper bounds and overlap):
  1. [possible] Process and file overhead (OS or security layer) (up to +323)
     Fix: Check for antivirus/EDR exclusions for code folders.
  2. [observed] Background load during the run (unscored win)
     Fix: Quit or restart the busy processes (a daemon pinned near 100% for hours is a bug, not load).

The HTML report goes further. It places you next to the bundled reference machines, shows the evidence behind each cause, and lists how many points each test costs you. Here is an example report for a Windows PC with real-time antivirus.

Reading your score

Software around the hardware matters a lot. The reference is a desktop PC running Ubuntu in WSL2. The same PC, running natively on Windows with real-time antivirus and search indexing, scored less than a tenth of that.

A few things to keep in mind:

• Each cause is marked observed (a probe saw it), likely (a probe plus a matching pattern), or possible (a pattern only).
• Each fix shows the most it could add to your score. Fixes overlap, so the gains do not add up.
• Single runs vary by about 10 to 15%. Use --runs 3 when you want a number to compare.

What usually slows agents down

agentperf looks for these causes. Each one has shown up on a real machine.

• Antivirus and search indexing on Windows. Real-time scanning checks every process start and every file write.
• Code under /mnt/c in WSL. Creating 2,000 small files took 41.8 s there, and 0.024 s on the Linux filesystem.
• A slow shell profile. Codex starts a login shell for every command, so your dotfiles run every time.
• Background load. One stuck system process cut a laptop's score almost in half.
• Synced folders (iCloud, OneDrive, Dropbox), power-saving plans, and spinning disks.
• Large repos with git's fsmonitor turned off.

Commands

Add --fetch-tools when you compare machines. It downloads a pinned ripgrep build, so every machine searches with the same binary. Downloads are checked by SHA-256 and stored in ~/.agentperf/tools; your PATH is not touched.

agentperf remote my-mac --fetch-tools --clean
agentperf remote my-pc --fetch-tools
agentperf remote my-pc --wsl Ubuntu --fetch-tools

agentperf live --dry-run                  # shows what would run and spends nothing
agentperf live --claude-model sonnet

How it works

agentperf builds the same test project on every machine: 6,000 source-like files in 400 folders, about 49 MB. Then it times seven kinds of work.

The weights come from one developer's 6,416 real Claude Code tool calls. Most of them (79%) were shell commands, so process start counts the most.

The tests are built to be fair across operating systems. Writes must reach the disk on every OS, cold reads skip the file cache, and every machine uses the same ripgrep build. docs/METHODOLOGY.md explains the rules, the statistics, and the scoring, and docs/FAQ.md answers the usual objections.

Privacy

A raw result file contains your hostname, file paths, and process names, so do not post it as is. agentperf share writes an anonymized copy and lists everything it changed. See SECURITY.md.

Contributing

Contributions are welcome. You can add a benchmark, improve a diagnosis rule, or send an anonymized result from your machine to grow the reference set. CONTRIBUTING.md explains how. To run the tests:

python3 -m unittest discover -s tests -t .

Roadmap

• Publish to PyPI, so that pipx install agentperf works.
• More reference machines, especially Linux desktops and laptops.
• A result history on each machine, so you can compare before and after a change.

License

MIT. agentperf is not affiliated with other projects that have a similar name.