Benchmarks
Performance is a release gate here, not a marketing page. Every number below is
generated at build time from the reports named in
docs/acceptance/performance-report.json, so this page cannot quietly pick up a
newer failed or incomplete run.
MiB/s throughput , and higher is better. p95
Matched 1,000-file CLI comparison
ESLint, official Oxlint, and OXC for TSRX lint the same byte-identical TSX files, with one no-debugger rule, through each tool’s own npm entry point, over 5 warmups and 20 measured processes. The separate mixed-file-types row is a paired internal workload rather than a cross-tool comparison, and it is the only lane that enters the native TSRX path. Report: benchmarks/comparative/results-1789846911273.json (2026-09-19), every budget passed.
×
ms
budget ≤ 1.250×
official Oxlint
OXC for TSRX, all-TSX lane
0
0.3
0.6
0.9
1.2
1.5
0
20
40
60
80
budget
OXC for TSRX / official Oxlint (matched TSX lane)
asserted ratio 1.206× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.250×
official Oxlint
OXC for TSRX, all-TSX lane
0
0.3
0.6
0.9
1.2
1.5
0
20
40
60
80
budget
OXC for TSRX / official Oxlint (matched TSX lane)
asserted ratio 1.206× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≥ 2×
OXC for TSRX, all-TSX lane
ESLint + typescript-eslint
0
3
6
9
12
15
0
200
400
600
800
1000
budget
ESLint / OXC for TSRX (matched TSX lane)
asserted ratio 13.177× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≥ 2×
OXC for TSRX, all-TSX lane
ESLint + typescript-eslint
0
3
6
9
12
15
0
200
400
600
800
1000
budget
ESLint / OXC for TSRX (matched TSX lane)
asserted ratio 13.177× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.500×
all-TSX product workload
mixed file types workload
0
0.5
1
1.5
2
0
20
40
60
80
100
120
budget
Paired mixed-file-types / all-TSX product workload
asserted ratio 1.409× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.500×
all-TSX product workload
mixed file types workload
0
0.5
1
1.5
2
0
20
40
60
80
100
120
budget
Paired mixed-file-types / all-TSX product workload
asserted ratio 1.409× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
Boundary Result Budget Status OXC for TSRX / official Oxlint (matched TSX lane) 1.206 × ≤ 1.250 × ✓ pass ESLint / OXC for TSRX (matched TSX lane) 13.177 × ≥ 2 × ✓ pass Paired mixed-file-types / all-TSX product workload 1.409 × ≤ 1.500 × ✓ pass
Native lint
Frozen release gate: throughput, same-build ordinary-path overhead, cold start, memory, and batch invariants. The npm-launcher ratio is a diagnostic boundary, not a speed claim. Report: benchmarks/native-lint/results-1789846859997.json (2026-09-19), every budget passed.
19 budgets, every budget passed. Show the numbers and the per-gate charts
Boundary Result Budget Status P01 median standard-path latency ratio 1.001 ≤ 1.050 ✓ pass P01 p95 standard-path latency ratio 1.005 ≤ 1.080 ✓ pass P01 diagnostic parity 1 exactly 1 ✓ pass P01 direct standard bypass 1 exactly 1 ✓ pass P02 median scan+copy+parse throughput 216.30 ≥ 75 ✓ pass P02 p95 scan+copy+parse throughput 201.43 ≥ 60 ✓ pass P02 equivalent-TSX throughput ratio 0.491 ≥ 0.450 ✓ pass P02 warm 10 KiB p95 scan+copy+parse latency 0.048 ≤ 1 ✓ pass P03 in-process one-thread lint throughput 110.39 ≥ 35 ✓ pass P03 CLI one-thread lint throughput 67.841 ≥ 35 ✓ pass P03 end-to-end CLI latency ratio 1.221 ≤ 1.350 ✓ pass P03 diagnostic parity 1 exactly 1 ✓ pass P03 config compiled once 1 exactly 1 ✓ pass P03 config one parse per file 1 exactly 1 ✓ pass P03 configured rule applied 1 exactly 1 ✓ pass P05 fresh-process TSRX p95 latency 3.882 ≤ 50 ✓ pass Direct Rust / official Oxlint npm-launcher p95 ratio 0.108 ≤ 1.250 ✓ pass P05 startup-rule diagnostic parity 1 exactly 1 ✓ pass P07 TSRX peak RSS 231030784 ≤ 250626048 ✓ pass
×
ms
budget ≤ 1.050×
same-build canonical control
product standard path
0
0.2
0.4
0.6
0.8
1
1.2
0
2
4
6
8
10
12
budget
P01 median standard-path latency ratio
asserted ratio 1.001× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.050×
same-build canonical control
product standard path
0
0.2
0.4
0.6
0.8
1
1.2
0
2
4
6
8
10
12
budget
P01 median standard-path latency ratio
asserted ratio 1.001× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.080×
same-build canonical control
product standard path
0
0.3
0.6
0.9
1.2
1.5
0
2
4
6
8
10
12
budget
P01 p95 standard-path latency ratio
asserted ratio 1.005× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.080×
same-build canonical control
product standard path
0
0.3
0.6
0.9
1.2
1.5
0
2
4
6
8
10
12
budget
P01 p95 standard-path latency ratio
asserted ratio 1.005× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
MiB/s
MiB/s
budget ≥ 75 MiB/s
all 30 samples
0
50
100
150
200
250
190
200
210
220
230
budget
P02 median scan+copy+parse throughput
measured 216.30 MiB/s · pass · below: 30 samples, median 216.30 MiB/s (solid line), p95 201.43 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 75 MiB/s
all 30 samples
0
50
100
150
200
250
190
200
210
220
230
budget
P02 median scan+copy+parse throughput
measured 216.30 MiB/s · pass · below: 30 samples, median 216.30 MiB/s (solid line), p95 201.43 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 60 MiB/s
all 30 samples
0
50
100
150
200
250
190
200
210
220
230
budget
P02 p95 scan+copy+parse throughput
measured 201.43 MiB/s · pass · below: 30 samples, median 216.30 MiB/s (solid line), p95 201.43 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 60 MiB/s
all 30 samples
0
50
100
150
200
250
190
200
210
220
230
budget
P02 p95 scan+copy+parse throughput
measured 201.43 MiB/s · pass · below: 30 samples, median 216.30 MiB/s (solid line), p95 201.43 MiB/s (dotted line)
×
ms
budget ≥ 0.450×
equivalent TSX parse
TSRX scan, copy, and parse
0
0.1
0.2
0.3
0.4
0.5
0.6
0
1
2
3
4
5
6
budget
P02 equivalent-TSX throughput ratio
asserted ratio 0.491× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≥ 0.450×
equivalent TSX parse
TSRX scan, copy, and parse
0
0.1
0.2
0.3
0.4
0.5
0.6
0
1
2
3
4
5
6
budget
P02 equivalent-TSX throughput ratio
asserted ratio 0.491× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
ms
ms
budget ≤ 1 ms
all 100 samples
0
0.2
0.4
0.6
0.8
1
1.2
0.04
0.06
0.08
0.1
budget
P02 warm 10 KiB p95 scan+copy+parse latency
measured 0.048 ms · pass · below: 100 samples, median 0.047 ms (solid line), p95 0.048 ms (dotted line)
ms
ms
budget ≤ 1 ms
all 100 samples
0
0.2
0.4
0.6
0.8
1
1.2
0.04
0.06
0.08
0.1
budget
P02 warm 10 KiB p95 scan+copy+parse latency
measured 0.048 ms · pass · below: 100 samples, median 0.047 ms (solid line), p95 0.048 ms (dotted line)
MiB/s
MiB/s
budget ≥ 35 MiB/s
all 30 samples
0
30
60
90
120
150
100
105
110
115
120
budget
P03 in-process one-thread lint throughput
measured 110.39 MiB/s · pass · below: 30 samples, median 110.39 MiB/s (solid line), p95 105.24 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 35 MiB/s
all 30 samples
0
30
60
90
120
150
100
105
110
115
120
budget
P03 in-process one-thread lint throughput
measured 110.39 MiB/s · pass · below: 30 samples, median 110.39 MiB/s (solid line), p95 105.24 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 35 MiB/s
all 30 samples
0
20
40
60
80
62.5
65
67.5
70
72.5
budget
P03 CLI one-thread lint throughput
measured 67.841 MiB/s · pass · below: 30 samples, median 67.841 MiB/s (solid line), p95 64.062 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 35 MiB/s
all 30 samples
0
20
40
60
80
62.5
65
67.5
70
72.5
budget
P03 CLI one-thread lint throughput
measured 67.841 MiB/s · pass · below: 30 samples, median 67.841 MiB/s (solid line), p95 64.062 MiB/s (dotted line)
×
ms
budget ≤ 1.350×
equivalent TSX CLI lint
TSRX CLI lint
0
0.5
1
1.5
2
0
5
10
15
20
budget
P03 end-to-end CLI latency ratio
asserted ratio 1.221× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.350×
equivalent TSX CLI lint
TSRX CLI lint
0
0.5
1
1.5
2
0
5
10
15
20
budget
P03 end-to-end CLI latency ratio
asserted ratio 1.221× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
ms
ms
budget ≤ 50 ms
all 20 samples
0
10
20
30
40
50
60
3
3.5
4
4.5
budget
P05 fresh-process TSRX p95 latency
measured 3.882 ms · pass · below: 20 samples, median 3.370 ms (solid line), p95 3.882 ms (dotted line)
ms
ms
budget ≤ 50 ms
all 20 samples
0
10
20
30
40
50
60
3
3.5
4
4.5
budget
P05 fresh-process TSRX p95 latency
measured 3.882 ms · pass · below: 20 samples, median 3.370 ms (solid line), p95 3.882 ms (dotted line)
×
ms
budget ≤ 1.250×
official Oxlint npm launcher
direct Rust TSRX process
0
0.3
0.6
0.9
1.2
1.5
0
10
20
30
40
budget
Direct Rust / official Oxlint npm-launcher p95 ratio
asserted ratio 0.108× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.250×
official Oxlint npm launcher
direct Rust TSRX process
0
0.3
0.6
0.9
1.2
1.5
0
10
20
30
40
budget
Direct Rust / official Oxlint npm-launcher p95 ratio
asserted ratio 0.108× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
MiB
MiB
budget ≤ 239.02 MiB
all 5 samples
0
50
100
150
200
250
300
220
221
222
223
budget
P07 TSRX peak RSS
measured 220.33 MiB · pass · below: 5 samples, median 220.33 MiB (solid line), p95 222.27 MiB (dotted line)
MiB
MiB
budget ≤ 239.02 MiB
all 5 samples
0
50
100
150
200
250
300
220
221
222
223
budget
P07 TSRX peak RSS
measured 220.33 MiB · pass · below: 5 samples, median 220.33 MiB (solid line), p95 222.27 MiB (dotted line)
Frozen release gate: formatter throughput, convergence scaling, memory, and batch invariants. The historical 16.6 MiB/s value is an absolute cross-corpus-derived floor, not a speedup claim. Report: benchmarks/native-format/results-1789846871499.json (2026-09-19), every budget passed.
25 budgets, every budget passed. Show the numbers and the per-gate charts
Boundary Result Budget Status p04 direct median ratio 1.211 ≤ 1.250 ✓ pass p04 direct p95 ratio 1.222 ≤ 1.350 ✓ pass p04 direct output parity 1 exactly 1 ✓ pass p04 direct bypass 1 exactly 1 ✓ pass p04 sequential median mib s 99.184 ≥ 15 ✓ pass p04 sequential p95 mib s 87.901 ≥ 12 ✓ pass Sequential throughput vs absolute 16.6 MiB/s floor 99.184 ≥ 16.600 ✓ pass p04 default thread mib s 322.35 ≥ 100 ✓ pass p04 generalized control median mib s 16.128 ≥ 12 ✓ pass p04 generalized control p95 mib s 14.897 ≥ 10 ✓ pass p04 generalized control linear scaling 1.009 ≤ 1.350 ✓ pass p04 generalized control idempotent 1 exactly 1 ✓ pass p04 generalized control one parse 1 exactly 1 ✓ pass p04 generalized dynamic style coverage 1 exactly 1 ✓ pass p04 generalized style metadata 1 exactly 1 ✓ pass p04 generalized no hidden embedded parse 1 exactly 1 ✓ pass p04 idempotent 1 exactly 1 ✓ pass p04 one parse 1 exactly 1 ✓ pass p04 config compiled once 1 exactly 1 ✓ pass p04 config one parse per file 1 exactly 1 ✓ pass p04 config options applied 1 exactly 1 ✓ pass p05 stdin p95 ms 3.364 ≤ 110 ✓ pass Direct Rust / official Oxfmt npm-launcher p95 ratio 0.101 ≤ 1.250 ✓ pass p05 complete output 1 exactly 1 ✓ pass p07 rss ratio 1.118 ≤ 1.300 ✓ pass
×
ms
budget ≤ 1.250×
canonical formatter control
product standard path
0
0.3
0.6
0.9
1.2
1.5
0
2
4
6
8
10
12
budget
p04 direct median ratio
asserted ratio 1.211× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.250×
canonical formatter control
product standard path
0
0.3
0.6
0.9
1.2
1.5
0
2
4
6
8
10
12
budget
p04 direct median ratio
asserted ratio 1.211× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.350×
canonical formatter control
product standard path
0
0.5
1
1.5
2
0
2
4
6
8
10
12
budget
p04 direct p95 ratio
asserted ratio 1.222× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.350×
canonical formatter control
product standard path
0
0.5
1
1.5
2
0
2
4
6
8
10
12
budget
p04 direct p95 ratio
asserted ratio 1.222× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
MiB/s
MiB/s
budget ≥ 15 MiB/s
all 30 samples
0
20
40
60
80
100
120
60
80
100
120
budget
p04 sequential median mib s
measured 99.184 MiB/s · pass · below: 30 samples, median 99.184 MiB/s (solid line), p95 87.901 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 15 MiB/s
all 30 samples
0
20
40
60
80
100
120
60
80
100
120
budget
p04 sequential median mib s
measured 99.184 MiB/s · pass · below: 30 samples, median 99.184 MiB/s (solid line), p95 87.901 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 12 MiB/s
all 30 samples
0
20
40
60
80
100
60
80
100
120
budget
p04 sequential p95 mib s
measured 87.901 MiB/s · pass · below: 30 samples, median 99.184 MiB/s (solid line), p95 87.901 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 12 MiB/s
all 30 samples
0
20
40
60
80
100
60
80
100
120
budget
p04 sequential p95 mib s
measured 87.901 MiB/s · pass · below: 30 samples, median 99.184 MiB/s (solid line), p95 87.901 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 16.600 MiB/s
all 30 samples
0
20
40
60
80
100
120
60
80
100
120
budget
Sequential throughput vs absolute 16.6 MiB/s floor
measured 99.184 MiB/s · pass · below: 30 samples, median 99.184 MiB/s (solid line), p95 87.901 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 16.600 MiB/s
all 30 samples
0
20
40
60
80
100
120
60
80
100
120
budget
Sequential throughput vs absolute 16.6 MiB/s floor
measured 99.184 MiB/s · pass · below: 30 samples, median 99.184 MiB/s (solid line), p95 87.901 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 100 MiB/s
all 15 samples
0
100
200
300
400
300
350
400
450
500
budget
p04 default thread mib s
measured 322.35 MiB/s · pass · below: 15 samples, median 423.56 MiB/s (solid line), p95 322.35 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 100 MiB/s
all 15 samples
0
100
200
300
400
300
350
400
450
500
budget
p04 default thread mib s
measured 322.35 MiB/s · pass · below: 15 samples, median 423.56 MiB/s (solid line), p95 322.35 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 12 MiB/s
all 15 samples
0
5
10
15
20
14
15
16
17
18
budget
p04 generalized control median mib s
measured 16.128 MiB/s · pass · below: 15 samples, median 16.128 MiB/s (solid line), p95 14.897 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 12 MiB/s
all 15 samples
0
5
10
15
20
14
15
16
17
18
budget
p04 generalized control median mib s
measured 16.128 MiB/s · pass · below: 15 samples, median 16.128 MiB/s (solid line), p95 14.897 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 10 MiB/s
all 15 samples
0
5
10
15
20
14
15
16
17
18
budget
p04 generalized control p95 mib s
measured 14.897 MiB/s · pass · below: 15 samples, median 16.128 MiB/s (solid line), p95 14.897 MiB/s (dotted line)
MiB/s
MiB/s
budget ≥ 10 MiB/s
all 15 samples
0
5
10
15
20
14
15
16
17
18
budget
p04 generalized control p95 mib s
measured 14.897 MiB/s · pass · below: 15 samples, median 16.128 MiB/s (solid line), p95 14.897 MiB/s (dotted line)
×
ms
budget ≤ 1.350×
half-size corpus
full generalized corpus
0
0.5
1
1.5
2
0
5
10
15
20
budget
p04 generalized control linear scaling
asserted ratio 1.009× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.350×
half-size corpus
full generalized corpus
0
0.5
1
1.5
2
0
5
10
15
20
budget
p04 generalized control linear scaling
asserted ratio 1.009× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
ms
ms
budget ≤ 110 ms
all 20 samples
0
30
60
90
120
150
2.5
2.75
3
3.25
3.5
3.75
budget
p05 stdin p95 ms
measured 3.364 ms · pass · below: 20 samples, median 2.926 ms (solid line), p95 3.364 ms (dotted line)
ms
ms
budget ≤ 110 ms
all 20 samples
0
30
60
90
120
150
2.5
2.75
3
3.25
3.5
3.75
budget
p05 stdin p95 ms
measured 3.364 ms · pass · below: 20 samples, median 2.926 ms (solid line), p95 3.364 ms (dotted line)
×
ms
budget ≤ 1.250×
official Oxfmt npm launcher
direct Rust TSRX stdin
0
0.3
0.6
0.9
1.2
1.5
0
10
20
30
40
budget
Direct Rust / official Oxfmt npm-launcher p95 ratio
asserted ratio 0.101× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.250×
official Oxfmt npm launcher
direct Rust TSRX stdin
0
0.3
0.6
0.9
1.2
1.5
0
10
20
30
40
budget
Direct Rust / official Oxfmt npm-launcher p95 ratio
asserted ratio 0.101× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
MiB
budget ≤ 1.300×
canonical TSX RSS
TSRX formatter RSS
0
0.3
0.6
0.9
1.2
1.5
0
50
100
150
200
250
budget
p07 rss ratio
asserted ratio 1.118× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
MiB
budget ≤ 1.300×
canonical TSX RSS
TSRX formatter RSS
0
0.3
0.6
0.9
1.2
1.5
0
50
100
150
200
250
budget
p07 rss ratio
asserted ratio 1.118× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
Opt-in type-aware lint
The opt-in TypeScript-Go lane. The default lane stays syntax-only with zero type processes. Report: benchmarks/type-aware/results-1789846873134.json (2026-09-19), every budget passed.
8 budgets, every budget passed. Show the numbers and the per-gate charts
Boundary Result Budget Status Default syntax lint p95 3.449 ms ≤ 10 ms ✓ pass Single-file type-aware p95 27.179 ms ≤ 60 ms ✓ pass Two-file project type-aware p95 27.933 ms ≤ 70 ms ✓ pass Single-file type-aware cold start 69.869 ms ≤ 100 ms ✓ pass Two-file project cold start 27.410 ms ≤ 120 ms ✓ pass Type-aware vs default p95 ratio 7.880 × ≤ 12 × ✓ pass Type processes per batch 1 exactly 1 ✓ pass Default path parse count per file 1 exactly 1 ✓ pass
ms
ms
budget ≤ 10 ms
all 20 samples
0
2
4
6
8
10
12
2
2.5
3
3.5
4
budget
Default syntax lint p95
measured 3.449 ms · pass · below: 20 samples, median 2.845 ms (solid line), p95 3.449 ms (dotted line)
ms
ms
budget ≤ 10 ms
all 20 samples
0
2
4
6
8
10
12
2
2.5
3
3.5
4
budget
Default syntax lint p95
measured 3.449 ms · pass · below: 20 samples, median 2.845 ms (solid line), p95 3.449 ms (dotted line)
ms
ms
budget ≤ 60 ms
all 20 samples
0
20
40
60
80
22
24
26
28
30
budget
Single-file type-aware p95
measured 27.179 ms · pass · below: 20 samples, median 25.227 ms (solid line), p95 27.179 ms (dotted line)
ms
ms
budget ≤ 60 ms
all 20 samples
0
20
40
60
80
22
24
26
28
30
budget
Single-file type-aware p95
measured 27.179 ms · pass · below: 20 samples, median 25.227 ms (solid line), p95 27.179 ms (dotted line)
ms
ms
budget ≤ 70 ms
all 20 samples
0
20
40
60
80
24
26
28
30
32
budget
Two-file project type-aware p95
measured 27.933 ms · pass · below: 20 samples, median 25.588 ms (solid line), p95 27.933 ms (dotted line)
ms
ms
budget ≤ 70 ms
all 20 samples
0
20
40
60
80
24
26
28
30
32
budget
Two-file project type-aware p95
measured 27.933 ms · pass · below: 20 samples, median 25.588 ms (solid line), p95 27.933 ms (dotted line)
ms
budget ≤ 100 ms
0
20
40
60
80
100
120
budget
Single-file type-aware cold start
single measurement per report: 69.869 ms · pass
ms
budget ≤ 100 ms
0
20
40
60
80
100
120
budget
Single-file type-aware cold start
single measurement per report: 69.869 ms · pass
ms
budget ≤ 120 ms
0
30
60
90
120
150
budget
Two-file project cold start
single measurement per report: 27.410 ms · pass
ms
budget ≤ 120 ms
0
30
60
90
120
150
budget
Two-file project cold start
single measurement per report: 27.410 ms · pass
×
ms
budget ≤ 12×
default syntax lane
single-file type-aware lane
0
3
6
9
12
15
0
10
20
30
40
budget
Type-aware vs default p95 ratio
asserted ratio 7.880× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 12×
default syntax lane
single-file type-aware lane
0
3
6
9
12
15
0
10
20
30
40
budget
Type-aware vs default p95 ratio
asserted ratio 7.880× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
Vite/Vite+ command boundary
Fresh companion processes at the ecosystem seam. Vite build and HMR carry zero OXC for TSRX transforms. Report: benchmarks/vite/results-1789846886123.json (2026-09-19), every budget passed.
8 budgets, every budget passed. Show the numbers and the per-gate charts
Boundary Result Budget Status Mixed companion lint p95 65.685 ms ≤ 150 ms ✓ pass Mixed lint vs canonical p95 ratio 1.926 × ≤ 2.500 × ✓ pass Mixed companion format-check p95 66.569 ms ≤ 220 ms ✓ pass Mixed format vs canonical p95 ratio 2.024 × ≤ 2.100 × ✓ pass Ordinary npm formatter p95 36.928 ms ≤ 150 ms ✓ pass Ordinary npm formatter vs canonical p95 ratio 1.123 × ≤ 1.250 × ✓ pass Vite+ 0.2.4 mixed lint p95 296.54 ms ≤ 750 ms ✓ pass Native TSRX parses per file 1 exactly 1 ✓ pass
ms
ms
budget ≤ 150 ms
all 20 samples
0
50
100
150
200
50
60
70
80
budget
Mixed companion lint p95
measured 65.685 ms · pass · below: 20 samples, median 62.336 ms (solid line), p95 65.685 ms (dotted line)
ms
ms
budget ≤ 150 ms
all 20 samples
0
50
100
150
200
50
60
70
80
budget
Mixed companion lint p95
measured 65.685 ms · pass · below: 20 samples, median 62.336 ms (solid line), p95 65.685 ms (dotted line)
×
ms
budget ≤ 2.500×
canonical lint
mixed companion lint
0
0.5
1
1.5
2
2.5
3
0
20
40
60
80
100
budget
Mixed lint vs canonical p95 ratio
asserted ratio 1.926× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 2.500×
canonical lint
mixed companion lint
0
0.5
1
1.5
2
2.5
3
0
20
40
60
80
100
budget
Mixed lint vs canonical p95 ratio
asserted ratio 1.926× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
ms
ms
budget ≤ 220 ms
all 20 samples
0
50
100
150
200
250
55
60
65
70
75
budget
Mixed companion format-check p95
measured 66.569 ms · pass · below: 20 samples, median 61.783 ms (solid line), p95 66.569 ms (dotted line)
ms
ms
budget ≤ 220 ms
all 20 samples
0
50
100
150
200
250
55
60
65
70
75
budget
Mixed companion format-check p95
measured 66.569 ms · pass · below: 20 samples, median 61.783 ms (solid line), p95 66.569 ms (dotted line)
×
ms
budget ≤ 2.100×
canonical format
mixed companion format
0
0.5
1
1.5
2
2.5
0
20
40
60
80
budget
Mixed format vs canonical p95 ratio
asserted ratio 2.024× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 2.100×
canonical format
mixed companion format
0
0.5
1
1.5
2
2.5
0
20
40
60
80
budget
Mixed format vs canonical p95 ratio
asserted ratio 2.024× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
ms
ms
budget ≤ 150 ms
all 20 samples
0
50
100
150
200
30
40
50
60
budget
Ordinary npm formatter p95
measured 36.928 ms · pass · below: 20 samples, median 35.609 ms (solid line), p95 36.928 ms (dotted line)
ms
ms
budget ≤ 150 ms
all 20 samples
0
50
100
150
200
30
40
50
60
budget
Ordinary npm formatter p95
measured 36.928 ms · pass · below: 20 samples, median 35.609 ms (solid line), p95 36.928 ms (dotted line)
×
ms
budget ≤ 1.250×
canonical format
ordinary npm formatter
0
0.3
0.6
0.9
1.2
1.5
0
10
20
30
40
50
60
budget
Ordinary npm formatter vs canonical p95 ratio
asserted ratio 1.123× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
×
ms
budget ≤ 1.250×
canonical format
ordinary npm formatter
0
0.3
0.6
0.9
1.2
1.5
0
10
20
30
40
50
60
budget
Ordinary npm formatter vs canonical p95 ratio
asserted ratio 1.123× · pass · below: the two runs it divides, sampled independently, so no per-sample ratios
ms
ms
budget ≤ 750 ms
all 20 samples
0
200
400
600
800
1000
260
280
300
320
budget
Vite+ 0.2.4 mixed lint p95
measured 296.54 ms · pass · below: 20 samples, median 277.05 ms (solid line), p95 296.54 ms (dotted line)
ms
ms
budget ≤ 750 ms
all 20 samples
0
200
400
600
800
1000
260
280
300
320
budget
Vite+ 0.2.4 mixed lint p95
measured 296.54 ms · pass · below: 20 samples, median 277.05 ms (solid line), p95 296.54 ms (dotted line)
Native editor server
Syntax-only local language-server round trips on the retained Markless fixture. Report: benchmarks/editor/results-1789846886872.json (2026-09-19), every budget passed.
6 budgets, every budget passed. Show the numbers and the per-gate charts
Boundary Result Budget Status Server start to first diagnostics 3.857 ms ≤ 50 ms ✓ pass Edit-to-diagnostics p95 0.169 ms ≤ 25 ms ✓ pass Formatting p95 0.138 ms ≤ 15 ms ✓ pass Safe code-action p95 0.155 ms ≤ 25 ms ✓ pass RSS after 1,000-edit soak 10.625 MiB ≤ 96 MiB ✓ pass RSS growth through soak 0 MiB ≤ 8 MiB ✓ pass
ms
budget ≤ 50 ms
0
10
20
30
40
50
60
budget
Server start to first diagnostics
single measurement per report: 3.857 ms · pass
ms
budget ≤ 50 ms
0
10
20
30
40
50
60
budget
Server start to first diagnostics
single measurement per report: 3.857 ms · pass
ms
ms
budget ≤ 25 ms
all 100 samples
0
5
10
15
20
25
30
0
0.1
0.2
0.3
0.4
budget
Edit-to-diagnostics p95
measured 0.169 ms · pass · below: 100 samples, median 0.097 ms (solid line), p95 0.169 ms (dotted line)
ms
ms
budget ≤ 25 ms
all 100 samples
0
5
10
15
20
25
30
0
0.1
0.2
0.3
0.4
budget
Edit-to-diagnostics p95
measured 0.169 ms · pass · below: 100 samples, median 0.097 ms (solid line), p95 0.169 ms (dotted line)
ms
ms
budget ≤ 15 ms
all 100 samples
0
5
10
15
20
0.05
0.1
0.15
0.2
0.25
0.3
budget
Formatting p95
measured 0.138 ms · pass · below: 100 samples, median 0.108 ms (solid line), p95 0.138 ms (dotted line)
ms
ms
budget ≤ 15 ms
all 100 samples
0
5
10
15
20
0.05
0.1
0.15
0.2
0.25
0.3
budget
Formatting p95
measured 0.138 ms · pass · below: 100 samples, median 0.108 ms (solid line), p95 0.138 ms (dotted line)
ms
ms
budget ≤ 25 ms
all 100 samples
0
5
10
15
20
25
30
0
0.1
0.2
0.3
0.4
budget
Safe code-action p95
measured 0.155 ms · pass · below: 100 samples, median 0.105 ms (solid line), p95 0.155 ms (dotted line)
ms
ms
budget ≤ 25 ms
all 100 samples
0
5
10
15
20
25
30
0
0.1
0.2
0.3
0.4
budget
Safe code-action p95
measured 0.155 ms · pass · below: 100 samples, median 0.105 ms (solid line), p95 0.155 ms (dotted line)
MiB
budget ≤ 96 MiB
0
20
40
60
80
100
120
budget
RSS after 1,000-edit soak
single measurement per report: 10.625 MiB · pass
MiB
budget ≤ 96 MiB
0
20
40
60
80
100
120
budget
RSS after 1,000-edit soak
single measurement per report: 10.625 MiB · pass
MiB
budget ≤ 8 MiB
0
2
4
6
8
10
budget
RSS growth through soak
single measurement per report: 0 MiB · pass
MiB
budget ≤ 8 MiB
0
2
4
6
8
10
budget
RSS growth through soak
single measurement per report: 0 MiB · pass
How a number gets on this page
Noise is handled by a fail-closed rerun policy rather than by picking a
favorable sample. A first report inside the 3% near-threshold band requires
exactly two additional fresh reports, and only the assertions that triggered get
two-of-three tolerance; everything else has to pass in all three. The
representative is chosen by median normalized budget pressure with a stable
report-path tie-break, never by taking the fastest run, and the aggregate fails
if the selected representative is red or if any failure lands more than 3%
beyond its threshold.
The harnesses are release-only, and memory-stats is linked into benchmark
executables only, never into the CLI you install. Make the authoritative release
decision with:
node tests/acceptance/run-performance.mjs
OXC for TSRX