[jdx]

🤖 Refreshed benchmarks

benchmarks/results.json was pinned to aube 1.19.0; the workspace is now 1.21.0. Re-ran mise run bench:bump on the hermetic Verdaccio registry (500mbit / 50ms per the mise task) and regenerated benchmarks/results.json plus the README BENCH_RATIOS block. The benchmark matrix pins aube's GVS mode via npm_config_enable_global_virtual_store=true|false (the auto-synthesized env alias for the enableGlobalVirtualStore setting), so GitHub Actions' inherited CI=true environment does not change whether aube runs with GVS enabled or disabled.

Benchmark changes

Versions:

• aube: 1.19.0 -> 1.21.0

Public ratios: warm installs vs Bun 6x -> 2x; warm installs vs pnpm 8x -> 3x; repeat test vs Bun 5x -> 5x; repeat test vs pnpm 34x -> 30x.

Benchmark	aube	bun	deno	pnpm	npm	yarn
Fresh install (warm cache)	308ms -> 886ms (+188%)	1889ms -> 1813ms (-4%)	1254ms -> 1616ms (+29%)	2515ms -> 2512ms (0%)	7080ms -> 7136ms (+1%)	8282ms -> 7708ms (-7%)
Fresh install (cold cache)	6209ms -> 5860ms (-6%)	5205ms -> 4856ms (-7%)	8135ms -> 6748ms (-17%)	13454ms -> 12925ms (-4%)	12234ms -> 8758ms (-28%)	14123ms -> 11079ms (-22%)
npm install && npm run test	12ms -> 11ms (-8%)	54ms -> 60ms (+11%)	95ms -> 59ms (-38%)	405ms -> 331ms (-18%)	780ms -> 681ms (-13%)	1351ms -> 1158ms (-14%)

Review the numbers before merging — if anything looks wildly off vs. the previous release, investigate before landing. Hermetic proxy jitter or an npmjs uplink hiccup can occasionally skew results.

Once merged to main, the updated bench results flow into the next release-plz-pr run automatically.

---

Generated by the bench-refresh workflow.

---

Note

Low Risk
Documentation and benchmark artifact updates only; no runtime or application code changes.

Overview
Refreshes published benchmark data for aube 1.21.0 (from 1.19.0) by updating benchmarks/results.json and the auto-generated BENCH_RATIOS block in README.md.

README marketing ratios are revised to match the new numbers: warm install vs pnpm 8x → 3x, vs Bun 6x → 2x, and repeat test vs pnpm 34x → 30x (vs Bun stays 5x).

The JSON run records new timings and stats for all three scenarios (warm/cold fresh install, install+test) across aube, bun, deno, pnpm, npm, and yarn—notably aube warm install rises from ~308ms to ~886ms while cold install and repeat-test numbers stay in a similar ballpark.

^{Reviewed by Cursor Bugbot for commit 63eae44. Bugbot is set up for automated code reviews on this repo. Configure here.}

Summary by CodeRabbit

• Documentation
  • Updated benchmark performance claims with revised measurements
  • Refreshed benchmark test results and version references

[coderabbitai]

📝 Walkthrough

Walkthrough

Benchmark results in benchmarks/results.json are refreshed for aube v1.21.0: the timestamp, version string, and all per-manager values and stats fields for the gvs-warm, gvs-cold, and install-test scenarios are replaced. The README.md warm-install and repeat-command speedup multipliers are updated to match the new data.

Changes

Benchmark Data and README Refresh

Layer / File(s)	Summary
Benchmark results data update benchmarks/results.json	Updates the top-level updated timestamp, bumps the recorded aube version from 1.19.0 to 1.21.0, and replaces all values arrays and recomputed stats (mean, stddev, min, max, text) for gvs-warm, gvs-cold, and install-test across all six package managers.
README speedup claims update README.md	Lowers the warm-install speedup ratios (pnpm: 8x→3x, Bun: 6x→2x) and adjusts the pnpm repeat-command multiplier (34x→30x) to reflect the new benchmark measurements.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Possibly related PRs

• jdx/aube#820: Updates the same README.md "Fast installs" ratios and benchmarks/results.json scenario rows (gvs-warm, gvs-cold, install-test).
• jdx/aube#851: Identical benchmark refresh pattern — aube version bump, recalculated stats for all three scenarios, and README multiplier updates.
• jdx/aube#866: Directly overlapping structure: regenerates benchmarks/results.json entries and "Fast installs" ratios in README.md.

Poem

🐇 Hop hop, the numbers changed today,
The benchmarks ran and led the way.
Three-ex warm, not eight — that's fair,
Thirty times repeat, still debonair.
v1.21 hops into view,
Fresh results for me and you! 🎉

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title 'chore: refresh benchmarks for v1.21.0' directly and specifically describes the main purpose of the pull request, which is to update benchmark results for aube v1.21.0.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[greptile-apps]

Greptile Summary

This PR refreshes benchmarks/results.json from aube 1.19.0 to 1.21.0 and regenerates the README's BENCH_RATIOS block accordingly. All values-to-stats consistency checks out, and the stated ratios (2x/3x warm, 5x/30x repeat) are arithmetically correct.

• aube's warm-cache install time regressed +188% (308ms → 886ms), which is acknowledged in the PR description — but merits investigation before landing since pnpm and bun were nearly flat.
• npm's warm-install stddev exploded 28x (0.041s → 1.138s), suggesting system noise during that benchmark run, which could contaminate other results in the same scenario.

Confidence Score: 3/5

The README ratios are arithmetically correct, but the warm-install benchmark environment shows signs of interference — npm's stddev spiked 28x and aube's warm-install time nearly tripled. Landing these numbers as the canonical benchmark baseline without a re-run carries some risk of publishing misleading perf claims.

aube's own benchmark jumped +188% in the warm-cache scenario, and npm's stddev explosion in the same scenario strongly suggests the benchmark machine was under load. If the environment was noisy, the 308ms→886ms aube result may not reflect actual 1.21.0 performance, which would make the published ratios inaccurate.

benchmarks/results.json — the gvs-warm scenario warrants a targeted re-run before these numbers are treated as canonical.

Important Files Changed

Filename	Overview
benchmarks/results.json	Benchmark data updated for aube 1.21.0; all values/stats are internally consistent, but aube's warm-install mean jumped +188% (308ms→886ms) and npm's warm-install stddev spiked 28x (0.041s→1.138s), signaling a potentially noisy benchmark environment for that scenario.
README.md	BENCH_RATIOS block updated; ratios (2x Bun, 3x pnpm warm; 5x Bun, 30x pnpm repeat) are arithmetically correct against the new results.json values.

_{Reviews (1): Last reviewed commit: "chore: refresh benchmarks" | Re-trigger Greptile}

[greptile-apps]

npm warm-install stddev spiked 28x — possible noisy environment

npm's gvs-warm stddev went from 0.041s to 1.138s (a 28x increase), with a min/max spread of 6.36s–8.82s vs the previous run's tight 7.02s–7.12s range. That level of variance strongly suggests OS or network interference during this scenario's run. A noisy environment can silently drag other tools' numbers in either direction too. Worth re-running the warm-cache scenario in isolation to confirm the aube 308ms→886ms jump is real and not an artifact of a loaded machine.

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!