Do coding agents recommend JMH?
JMH is the most chosen performance testing tool, taking 8% of 359 judged performance testing sessions.
JMH is the most chosen performance testing tool, taking 8% of 359 judged performance testing sessions. It was also raised as a candidate in 25 further sessions without being chosen.
This page reports what happened when Claude Code, Codex, Cursor, Grok Build CLI and Muse Code had to solve a problem in performance testing inside a realistic codebase. Not what a chat assistant says about JMH. 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
| Category | Performance testing |
| Sessions in the category | 359 |
| Sessions where JMH was chosen | 30 |
| Install share | 8% |
| Rank in category | 1 of 16 |
| Codebases it won in | 1 |
| Raised as a candidate, not chosen | 25 |
| Chosen when considered | 55% |
| Site | openjdk.org |
By agent
Claude Code, Codex, Cursor and Muse Code land within 7 points of each other on JMH, which is closer than most products in this experiment manage. Grok Build CLI ran fewer than 30 sessions here, too few to compare.
| Agent | Sessions | Chose JMH | Share |
|---|---|---|---|
| Claude Code | 89 | 8 | 9% |
| Codex | 108 | 12 | 11% |
| Cursor | 90 | 7 | 8% |
| Grok Build CLI | 18 | 1 | 6% |
| Muse Code | 54 | 2 | 4% |
By who was asking
JMH does much better with one kind of buyer than another. It won 30% of sessions asked as enterprise team and 0% of those asked as senior engineer.
| Who is asking | Sessions | Chose JMH | Share |
|---|---|---|---|
| Junior developer | 20 | 0 | 0% |
| Senior engineer | 239 | 0 | 0% |
| Enterprise team | 100 | 30 | 30% |
What JMH was up against
The full ranking in performance testing, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Built in-house (no product adopted) | 202 | 56% |
| 2 | JMH (this page) | 30 | 8% |
| 3 | pytest-benchmark | 22 | 6% |
| 4 | Grafana k6 | 22 | 6% |
| 5 | hyperfine | 19 | 5% |
| 6 | PHPBench | 19 | 5% |
| 7 | Autocannon | 19 | 5% |
| 8 | BenchmarkDotNet | 8 | 2% |
What this means
Leading a category is a position to defend rather than a result to celebrate. In our data, leaders lose ground in exactly two situations: when the person asking is an enterprise buyer with procurement constraints, and when a competitor's documentation is easier for an agent to integrate correctly.
The defence is unglamorous. Keep the quickstart running. Keep the version current on the page. Keep the names aligned. Stay in the templates.
Where these numbers come from
The 359 sessions in performance testing are part of a published set of 15,287, 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 JMH: 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 JMH?
Yes. JMH was chosen in 30 of the 359 judged sessions in performance testing, a 8% install share, ranking first in its category.
Does Claude Code recommend JMH?
In 8 of the 89 sessions in performance testing run with Claude Code, which is 9%.
Do different coding agents treat JMH differently?
Not much. Claude Code, Codex, Cursor and Muse Code chose it at similar rates, between 4% and 11% of their runs.
How was this measured?
Real coding agents at pinned versions were run in sandboxes inside 110 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 JMH considered but not chosen?
It was raised as a candidate in 25 sessions without being chosen, and chosen in 30. That is a 55% conversion from considered to chosen.
Where this comes from
Armature ran 15,287 judged sessions with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code inside 110 realistic codebases, and published every run. The numbers on this page come from that work.