chore: refresh benchmarks for v1.19.0

[jdx]

🤖 Refreshed benchmarks

benchmarks/results.json was pinned to aube 1.18.2; the workspace is now 1.19.0. 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: 1.18.2 -> 1.19.0
• deno: 2.8.2 -> 2.8.3
• pnpm: 11.5.2 -> 11.6.0

Public ratios: warm installs vs Bun 7x -> 6x; warm installs vs pnpm 6x -> 8x; repeat test vs Bun 5x -> 5x; repeat test vs pnpm 29x -> 34x.

Benchmark	aube	bun	deno	pnpm	npm	yarn
Fresh install (warm cache)	383ms -> 308ms (-20%)	2728ms -> 1889ms (-31%)	1329ms -> 1254ms (-6%)	2364ms -> 2515ms (+6%)	7196ms -> 7080ms (-2%)	8382ms -> 8282ms (-1%)
Fresh install (cold cache)	8132ms -> 6209ms (-24%)	5747ms -> 5205ms (-9%)	8815ms -> 8135ms (-8%)	23602ms -> 13454ms (-43%)	31075ms -> 12234ms (-61%)	18714ms -> 14123ms (-25%)
npm install && npm run test	13ms -> 12ms (-8%)	69ms -> 54ms (-22%)	83ms -> 95ms (+14%)	375ms -> 405ms (+8%)	1005ms -> 780ms (-22%)	1595ms -> 1351ms (-15%)

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 JSON only; no runtime or package-manager logic changes.

Overview
Refreshes published benchmark numbers for aube 1.19.0 (from 1.18.2) by re-running the hermetic bench:bump flow and committing benchmarks/results.json.

The README Fast installs blurb (BENCH_RATIOS) is regenerated from those results: warm install vs pnpm 6x → 8x, vs Bun 7x → 6x; repeat install && test vs pnpm 29x → 34x (vs Bun stays 5x). Tool pins in the JSON also move to deno 2.8.3 and pnpm 11.6.0, with faster aube times across warm/cold install and repeat-test scenarios alongside shifts for other managers (notably large cold-cache drops for npm/pnpm in this run).

^{Reviewed by Cursor Bugbot for commit 00fcb0d. Bugbot is set up for automated code reviews on this repo. Configure here.}

Summary by CodeRabbit

Documentation

• Updated benchmark performance metrics with recent test results.
• Refreshed tool versions in the benchmark suite.

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR updates benchmark data and documentation for the Aube package manager. The benchmark results file is refreshed with new measurement data across three benchmark suites, tool versions are bumped, and the README is updated to display the new performance ratios.

Changes

Benchmark Results Update

Layer / File(s)	Summary
Benchmark measurement data and metadata benchmarks/results.json	Updated timestamp and bumped Aube 1.18.2 → 1.19.0, Deno 2.8.2 → 2.8.3, and pnpm 11.5.2 → 11.6.0 in the versions section. Replaced all benchmark measurement values and statistics (mean, stddev, min, max) for gvs-warm, gvs-cold, and install-test rows.
README benchmark documentation README.md	Updated the "Fast installs" section with new warm install speedup and repeat test command speedup ratios vs pnpm and Bun to match the refreshed benchmark data.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~5 minutes

Possibly related PRs

• jdx/aube#841: Updates the same benchmark artifacts with refreshed measurement data for gvs-warm, gvs-cold, install-test rows and README ratios.
• jdx/aube#851: Refreshes benchmarks/results.json and README.md "Fast installs" ratios with new performance measurements.
• jdx/aube#820: Updates both benchmark data file and README with new performance test results and version bumps.

Poem

🐰 A hop through numbers, fresh and bright—
New benchmarks glow in measured light,
The speedups bloom where versions grow,
From data runs to docs they flow.
Aube hops faster, we now show! ⚡

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title 'chore: refresh benchmarks for v1.19.0' clearly and concisely describes the main change—updating benchmark results for the v1.19.0 release of aube.
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 benchmark data from aube 1.18.2 → 1.19.0 (along with deno 2.8.2 → 2.8.3 and pnpm 11.5.2 → 11.6.0) and updates the README marketing ratios to match.

• benchmarks/results.json: All three benchmark scenarios (gvs-warm, gvs-cold, install-test) updated with new timing values and per-manager stats; the rounded integer values fields, floating-point mean/stddev, and human-readable text strings are all mutually consistent.
• README.md: The BENCH_RATIOS block is updated from the old ratios (6x pnpm, 7x Bun, 29x pnpm test, 5x Bun test) to the new ones (8x pnpm, 6x Bun, 34x pnpm test, 5x Bun test), which correctly round from the updated results.

Confidence Score: 5/5

Only benchmark data and README prose are changed — no code paths, logic, or configuration are affected.

Both files contain purely numerical benchmark results and derived marketing copy. The stated ratios in the README are arithmetically correct against the new results.json values, all stats fields (mean, stddev, min, max, text) are internally consistent, and version bumps match what is described in the PR. The one eyebrow-raiser — npm cold-cache dropping from ~31 s to ~12 s — is flagged by the author and is consistent with a tighter stddev in the new run (±0.31 s vs ±1.63 s), suggesting a cleaner hermetic environment rather than a measurement error.

No files require special attention.

Important Files Changed

Filename	Overview
benchmarks/results.json	Benchmark data refreshed for aube 1.19.0, deno 2.8.3, and pnpm 11.6.0; all values, stats, and text fields are internally consistent.
README.md	BENCH_RATIOS block updated to match the new results.json numbers; all stated ratios are mathematically correct.

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