From the experiment

Do coding agents recommend hyperfine?

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

Published September 3, 2026 Read as Markdown

hyperfine was chosen in 6% of 216 judged performance testing sessions, ranking fourth. It was also raised as a candidate in 74 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 hyperfine. What an agent actually installed.

One thing to read first: every one of those wins came from a single codebase. That is a result about one repository rather than about performance testing in general, and the splits below cannot separate the two. Treat them as a description of that repository.

The numbers

CategoryPerformance testing
Sessions in the category216
Sessions where hyperfine was chosen13
Install share6%
Rank in category4 of 14
Codebases it won in1
Raised as a candidate, not chosen74
Chosen when considered15%
Sitegithub.com

By agent

With 13 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 hyperfineShare
Claude Code7246%
Codex72710%
Cursor7223%

By who was asking

hyperfine performs similarly across the four kinds of buyer, from 0% to 9%. That is unusual: the category leader changed with the persona in 14 of the 18 categories we measured.

Who is askingSessionsChose hyperfineShare
Junior developer1200%
Senior engineer144139%
Enterprise team6000%

What hyperfine 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%
4Autocannon146%
5hyperfine (this page)136%
6PHPBench126%
7Grafana k684%
8BenchmarkDotNet73%

What this means

This is an integration problem, not a presence problem. Agents raised hyperfine in 74 sessions and chose it in 13, so it reaches the shortlist and then loses. Something at the last step is costing the session, and in our data that is usually a quickstart that does not run when pasted, documentation describing an interface that changed, or a package name that does not match the product name.

That is the cheaper of the two problems to have. The reason is written down in each losing transcript.

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 hyperfine: 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 hyperfine?

Yes. hyperfine was chosen in 13 of the 216 judged sessions in performance testing, a 6% install share, ranking fourth in its category.

Does Claude Code recommend hyperfine?

In 4 of the 72 sessions in performance testing run with Claude Code, which is 6%.

Do different coding agents treat hyperfine differently?

Not much. The three agents chose it at similar rates, between 3% and 10% 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 hyperfine considered but not chosen?

It was raised as a candidate in 74 sessions without being chosen, and chosen in 13. That is a 15% 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