From the experiment

Do coding agents recommend Autocannon?

Autocannon was chosen in 6% of 216 judged performance testing sessions, ranking third. Measured with Claude Code, Codex and Cursor.

Published September 3, 2026 Read as Markdown

Autocannon was chosen in 6% of 216 judged performance testing sessions, ranking third. It was also raised as a candidate in 9 further sessions without being chosen.

This page reports what happened when Claude Code, Codex and Cursor had to solve a problem in performance testing inside a realistic codebase. Not what a chat assistant says about Autocannon. What an agent actually installed.

The numbers

CategoryPerformance testing
Sessions in the category216
Sessions where Autocannon was chosen14
Install share6%
Rank in category3 of 14
Codebases it won in2
Raised as a candidate, not chosen9
Chosen when considered61%
Sitegithub.com

By agent

With 14 wins spread across three agents, the rates below are small numbers and a difference between them is not yet a finding. They are here because the direction is worth knowing, not because the gap is established.

AgentSessionsChose AutocannonShare
Claude Code7257%
Codex7257%
Cursor7246%

By who was asking

Autocannon performs similarly across the four kinds of buyer, from 0% to 10%. That is unusual: the category leader changed with the persona in 14 of the 18 categories we measured.

Who is askingSessionsChose AutocannonShare
Junior developer1200%
Senior engineer1441410%
Enterprise team6000%

What Autocannon was up against

The full ranking in performance testing, from the same sessions:

#ProductRuns wonShare
1Built in-house (no product adopted)11352%
2JMH2210%
3pytest-benchmark157%
4Autocannon (this page)146%
5hyperfine136%
6PHPBench126%
7Grafana k684%
8BenchmarkDotNet73%

What this means

Agents almost never raise Autocannon without choosing it: 9 sessions raised against 14 chosen. When it gets considered, it usually wins. The constraint is how rarely it gets considered at all, which is a presence problem: templates, framework integrations and pages that answer the configuration and production questions agents actually search for.

Where these numbers come from

The 216 sessions in performance testing are part of a published set of 5,292, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.

Every performance testing run can be replayed on the board.

If you work on Autocannon: the judge recorded a reason for every session where it was raised and passed over. Those reasons are in the transcripts.

<!-- generated by scripts/write-data-pages.mjs -->

Common questions

Do coding agents recommend Autocannon?

Yes. Autocannon was chosen in 14 of the 216 judged sessions in performance testing, a 6% install share, ranking third in its category.

Does Claude Code recommend Autocannon?

In 5 of the 72 sessions in performance testing run with Claude Code, which is 7%.

Do different coding agents treat Autocannon differently?

Not much. The three agents chose it at similar rates, between 6% and 7% of their runs.

How was this measured?

Real coding agents at pinned versions were run in sandboxes inside 51 realistic codebases and asked to solve real tasks. A simulated project owner approved or questioned each recommendation before any code was written, and a judge from a model family that builds none of the agents read every session blind.

How often is Autocannon considered but not chosen?

It was raised as a candidate in 9 sessions without being chosen, and chosen in 14. That is a 61% conversion from considered to chosen.

Where this comes from

Armature ran 5,292 judged sessions with Claude Code, Codex and Cursor inside 51 realistic codebases, and published every run. The numbers on this page come from that work.

Read next

All library pages