benchmark_latest

Jump to bottom

github-actions[bot] edited this page Aug 20, 2026 · 32 revisions

Latest known benchmark result, per method

Generated. Do not edit. Produced by .github/workflows/bench-nightly.yml from the wiki's own history, alongside nightly_run.

Not a comparison across methods. Each section below is the last night that method was actually re-run -- whichever night touched the source near it, not necessarily last night, and not the same night as its neighbours here. Every run measures on a GitHub hosted runner whose hardware differs night to night, so a number here says "this is the last known reading", never "faster than the section above it".

Per method

Lodestar.Text.Benchmarks.BatchEmbeddingBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	CorpusSize	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Allocated	Alloc Ratio
UnitLoop	1	6.055 μs	0.4960 μs	0.0272 μs	1.00	0.01	0.1678	-	2.76 KB	1.00
EmbedBatch	1	6.032 μs	0.0168 μs	0.0009 μs	1.00	0.00	0.1907	-	3.16 KB	1.14
EmbedBatchBucketed	1	5.970 μs	0.2114 μs	0.0116 μs	0.99	0.00	0.1907	-	3.16 KB	1.14

UnitLoop	8	109.842 μs	29.8926 μs	1.6385 μs	1.00	0.02	8.1787	0.2441	134.29 KB	1.00
EmbedBatch	8	68.536 μs	0.7210 μs	0.0395 μs	0.62	0.01	7.6904	0.3662	127.31 KB	0.95
EmbedBatchBucketed	8	71.211 μs	18.8124 μs	1.0312 μs	0.65	0.01	7.6904	0.3662	127.31 KB	0.95

UnitLoop	32	414.477 μs	8.7068 μs	0.4772 μs	1.00	0.00	28.3203	0.9766	468.71 KB	1.00
EmbedBatch	32	240.034 μs	8.7065 μs	0.4772 μs	0.58	0.00	26.8555	1.7090	441.32 KB	0.94
EmbedBatchBucketed	32	232.990 μs	112.9029 μs	6.1886 μs	0.56	0.01	26.1230	1.7090	427.8 KB	0.91

UnitLoop	128	1,583.137 μs	258.4457 μs	14.1663 μs	1.00	0.01	113.2813	5.8594	1874.78 KB	1.00
EmbedBatch	128	957.366 μs	167.9800 μs	9.2076 μs	0.60	0.01	107.4219	15.6250	1764.42 KB	0.94
EmbedBatchBucketed	128	933.548 μs	50.8695 μs	2.7883 μs	0.59	0.00	103.5156	15.6250	1696.9 KB	0.91

Lodestar.Text.Benchmarks.BpeBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Ratio	Gen0	Allocated	Alloc Ratio
Unigram	685.0 ms	100.86 ms	5.53 ms	1.00	32000.0000	519.51 MB	1.00
Bpe	550.6 ms	50.95 ms	2.79 ms	0.80	7000.0000	112.18 MB	0.22

Lodestar.Text.Benchmarks.BpeScalingBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Gen0	Gen1	Allocated
BpeOnOnePathologicalToken	512	103.4 μs	12.08 μs	0.66 μs	1.2207	-	20.38 KB
BpeOnOnePathologicalToken	1024	220.0 μs	52.01 μs	2.85 μs	2.4414	-	39.93 KB
BpeOnOnePathologicalToken	2048	481.2 μs	14.68 μs	0.80 μs	4.3945	-	78.98 KB
BpeOnOnePathologicalToken	4096	974.5 μs	35.83 μs	1.96 μs	8.7891	0.9766	157.03 KB

Lodestar.Text.Benchmarks.BucketRouteDiagnostics-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Allocated
DpGroup	12.75 μs	4.647 μs	0.255 μs	-
MyersGroup	134.22 μs	15.608 μs	0.856 μs	-

Lodestar.Text.Benchmarks.FuzzBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Allocated	Alloc Ratio
Ratio	98.92 ns	0.797 ns	0.044 ns	1.00	0.00	-	-	NA
PartialRatio	18,645.18 ns	3,578.370 ns	196.143 ns	188.49	1.72	-	-	NA
TokenSortRatio	1,048.15 ns	450.848 ns	24.713 ns	10.60	0.22	0.0782	1312 B	NA
TokenSetRatio	3,331.51 ns	683.035 ns	37.439 ns	33.68	0.33	0.3433	5760 B	NA
WRatio	4,503.32 ns	367.659 ns	20.153 ns	45.53	0.18	0.4272	7200 B	NA

Lodestar.Text.Benchmarks.IndelBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Distance_Utf16	8	26.79 ns	5.281 ns	0.289 ns	1.00	0.01	-	NA
Distance_CodePoint	8	135.35 ns	6.651 ns	0.365 ns	5.05	0.05	-	NA
NormalizedSimilarity_Utf16	8	27.33 ns	1.098 ns	0.060 ns	1.02	0.01	-	NA
SubsequenceLength_Utf16	8	26.67 ns	1.914 ns	0.105 ns	1.00	0.01	-	NA

Distance_Utf16	12	28.42 ns	4.088 ns	0.224 ns	1.00	0.01	-	NA
Distance_CodePoint	12	152.49 ns	20.265 ns	1.111 ns	5.37	0.05	-	NA
NormalizedSimilarity_Utf16	12	28.45 ns	1.243 ns	0.068 ns	1.00	0.01	-	NA
SubsequenceLength_Utf16	12	28.22 ns	0.875 ns	0.048 ns	0.99	0.01	-	NA

Distance_Utf16	16	31.00 ns	3.891 ns	0.213 ns	1.00	0.01	-	NA
Distance_CodePoint	16	173.45 ns	6.011 ns	0.329 ns	5.60	0.03	-	NA
NormalizedSimilarity_Utf16	16	32.13 ns	1.824 ns	0.100 ns	1.04	0.01	-	NA
SubsequenceLength_Utf16	16	30.73 ns	2.154 ns	0.118 ns	0.99	0.01	-	NA

Distance_Utf16	20	86.94 ns	9.304 ns	0.510 ns	1.00	0.01	-	NA
Distance_CodePoint	20	189.93 ns	7.121 ns	0.390 ns	2.18	0.01	-	NA
NormalizedSimilarity_Utf16	20	86.54 ns	2.145 ns	0.118 ns	1.00	0.01	-	NA
SubsequenceLength_Utf16	20	93.33 ns	5.656 ns	0.310 ns	1.07	0.01	-	NA

Distance_Utf16	24	55.92 ns	11.883 ns	0.651 ns	1.00	0.01	-	NA
Distance_CodePoint	24	632.62 ns	721.741 ns	39.561 ns	11.31	0.62	-	NA
NormalizedSimilarity_Utf16	24	54.46 ns	1.778 ns	0.097 ns	0.97	0.01	-	NA
SubsequenceLength_Utf16	24	61.10 ns	17.811 ns	0.976 ns	1.09	0.02	-	NA

Distance_Utf16	32	59.01 ns	1.482 ns	0.081 ns	1.00	0.00	-	NA
Distance_CodePoint	32	1,030.82 ns	102.558 ns	5.622 ns	17.47	0.09	-	NA
NormalizedSimilarity_Utf16	32	63.20 ns	1.224 ns	0.067 ns	1.07	0.00	-	NA
SubsequenceLength_Utf16	32	60.85 ns	2.134 ns	0.117 ns	1.03	0.00	-	NA

Distance_Utf16	128	1,288.99 ns	73.329 ns	4.019 ns	1.00	0.00	-	NA
Distance_CodePoint	128	24,036.85 ns	5,779.939 ns	316.818 ns	18.65	0.22	-	NA
NormalizedSimilarity_Utf16	128	1,414.86 ns	79.123 ns	4.337 ns	1.10	0.00	-	NA
SubsequenceLength_Utf16	128	1,381.37 ns	7.429 ns	0.407 ns	1.07	0.00	-	NA

Distance_Utf16	512	11,455.12 ns	1,100.840 ns	60.341 ns	1.00	0.01	-	NA
Distance_CodePoint	512	339,870.54 ns	13,900.829 ns	761.951 ns	29.67	0.15	-	NA
NormalizedSimilarity_Utf16	512	11,771.02 ns	413.154 ns	22.646 ns	1.03	0.00	-	NA
SubsequenceLength_Utf16	512	11,455.49 ns	355.239 ns	19.472 ns	1.00	0.00	-	NA

Lodestar.Text.Benchmarks.LcsGateBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Band	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Dp	8	135.02 ns	67.005 ns	3.673 ns	1.00	0.03	-	NA
Kernel	8	56.11 ns	0.754 ns	0.041 ns	0.42	0.01	-	NA

Dp	12	218.39 ns	40.015 ns	2.193 ns	1.00	0.01	-	NA
Kernel	12	61.40 ns	0.839 ns	0.046 ns	0.28	0.00	-	NA

Dp	14	296.35 ns	169.600 ns	9.296 ns	1.00	0.04	-	NA
Kernel	14	64.69 ns	4.839 ns	0.265 ns	0.22	0.01	-	NA

Dp	16	360.68 ns	49.661 ns	2.722 ns	1.00	0.01	-	NA
Kernel	16	69.54 ns	14.800 ns	0.811 ns	0.19	0.00	-	NA

Dp	18	439.26 ns	44.538 ns	2.441 ns	1.00	0.01	-	NA
Kernel	18	72.83 ns	1.452 ns	0.080 ns	0.17	0.00	-	NA

Dp	20	769.21 ns	34.819 ns	1.909 ns	1.00	0.00	-	NA
Kernel	20	77.46 ns	0.463 ns	0.025 ns	0.10	0.00	-	NA

Dp	24	1,016.34 ns	96.383 ns	5.283 ns	1.00	0.01	-	NA
Kernel	24	86.92 ns	2.117 ns	0.116 ns	0.09	0.00	-	NA

Dp	32	1,595.87 ns	206.523 ns	11.320 ns	1.00	0.01	-	NA
Kernel	32	103.46 ns	2.408 ns	0.132 ns	0.06	0.00	-	NA

Dp	48	3,124.78 ns	1,096.630 ns	60.110 ns	1.00	0.02	-	NA
Kernel	48	133.60 ns	4.769 ns	0.261 ns	0.04	0.00	-	NA

Dp	64	5,554.25 ns	671.180 ns	36.790 ns	1.00	0.01	-	NA
Kernel	64	159.30 ns	3.584 ns	0.196 ns	0.03	0.00	-	NA

Dp	96	13,123.09 ns	27,222.790 ns	1,492.173 ns	1.01	0.15	-	NA
Kernel	96	1,104.40 ns	29.709 ns	1.628 ns	0.08	0.01	-	NA

Lodestar.Text.Benchmarks.LevenshteinBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Ratio	Allocated	Alloc Ratio
Distance_Utf16	8	25.64 ns	0.361 ns	0.020 ns	1.00	-	NA
Distance_CodePoint	8	118.50 ns	2.243 ns	0.123 ns	4.62	-	NA
NormalizedSimilarity_Utf16	8	25.02 ns	1.381 ns	0.076 ns	0.98	-	NA

Distance_Utf16	64	267.29 ns	8.845 ns	0.485 ns	1.00	-	NA
Distance_CodePoint	64	660.06 ns	9.018 ns	0.494 ns	2.47	-	NA
NormalizedSimilarity_Utf16	64	268.91 ns	39.672 ns	2.175 ns	1.01	-	NA

Distance_Utf16	512	13,844.14 ns	160.635 ns	8.805 ns	1.00	-	NA
Distance_CodePoint	512	16,633.08 ns	410.864 ns	22.521 ns	1.20	-	NA
NormalizedSimilarity_Utf16	512	14,251.22 ns	200.249 ns	10.976 ns	1.03	-	NA

Lodestar.Text.Benchmarks.LevenshteinCodePointBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Distinct	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Distance_CodePoint	16	32	330.9 ns	4.16 ns	0.23 ns	1.00	0.00	-	NA
Distance_Utf16	16	32	1,442.2 ns	146.71 ns	8.04 ns	4.36	0.02	-	NA

Distance_CodePoint	16	512	333.9 ns	3.55 ns	0.19 ns	1.00	0.00	-	NA
Distance_Utf16	16	512	1,576.2 ns	834.02 ns	45.72 ns	4.72	0.12	-	NA

Distance_CodePoint	24	32	414.8 ns	38.50 ns	2.11 ns	1.00	0.01	-	NA
Distance_Utf16	24	32	3,501.3 ns	179.98 ns	9.87 ns	8.44	0.04	-	NA

Distance_CodePoint	24	512	420.6 ns	9.79 ns	0.54 ns	1.00	0.00	-	NA
Distance_Utf16	24	512	3,674.1 ns	186.98 ns	10.25 ns	8.74	0.02	-	NA

Distance_CodePoint	32	32	495.7 ns	4.67 ns	0.26 ns	1.00	0.00	-	NA
Distance_Utf16	32	32	6,398.1 ns	281.51 ns	15.43 ns	12.91	0.03	-	NA

Distance_CodePoint	32	512	501.4 ns	4.89 ns	0.27 ns	1.00	0.00	-	NA
Distance_Utf16	32	512	6,741.4 ns	1,841.26 ns	100.93 ns	13.44	0.17	-	NA

Distance_CodePoint	40	32	579.1 ns	2.82 ns	0.15 ns	1.00	0.00	-	NA
Distance_Utf16	40	32	10,209.6 ns	63.19 ns	3.46 ns	17.63	0.01	-	NA

Distance_CodePoint	40	512	615.9 ns	754.31 ns	41.35 ns	1.00	0.08	-	NA
Distance_Utf16	40	512	10,840.5 ns	1,738.88 ns	95.31 ns	17.65	1.00	-	NA

Distance_CodePoint	128	32	2,296.5 ns	349.50 ns	19.16 ns	1.00	0.01	-	NA
Distance_Utf16	128	32	128,741.7 ns	48,237.74 ns	2,644.07 ns	56.06	1.08	-	NA

Distance_CodePoint	128	512	2,353.2 ns	144.39 ns	7.91 ns	1.00	0.00	-	NA
Distance_Utf16	128	512	111,747.1 ns	17,073.87 ns	935.88 ns	47.49	0.37	-	NA

Distance_CodePoint	512	32	18,416.9 ns	432.52 ns	23.71 ns	1.00	0.00	-	NA
Distance_Utf16	512	32	2,348,294.3 ns	301,978.52 ns	16,552.46 ns	127.51	0.79	-	NA

Distance_CodePoint	512	512	445,933.0 ns	111,305.72 ns	6,101.04 ns	1.00	0.02	-	NA
Distance_Utf16	512	512	1,821,333.0 ns	36,225.45 ns	1,985.64 ns	4.08	0.05	-	NA

Lodestar.Text.Benchmarks.MetricsBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Samples	Classes	Mean	Error	StdDev	Gen0	Allocated
Matrix	1000	2	7,653.1 ns	1,317.85 ns	72.24 ns	0.0153	312 B
MatrixWeighted	1000	2	7,361.7 ns	608.87 ns	33.37 ns	0.0153	312 B
AccuracyScore	1000	2	919.6 ns	47.24 ns	2.59 ns	-	-
F1Macro	1000	2	7,640.7 ns	965.86 ns	52.94 ns	0.0229	472 B
Report	1000	2	10,417.8 ns	2,561.84 ns	140.42 ns	0.3815	6520 B
Matrix	1000	10	7,859.8 ns	908.55 ns	49.80 ns	0.0610	1248 B
MatrixWeighted	1000	10	7,595.3 ns	146.17 ns	8.01 ns	0.0687	1248 B
AccuracyScore	1000	10	918.1 ns	7.81 ns	0.43 ns	-	-
F1Macro	1000	10	8,163.3 ns	333.85 ns	18.30 ns	0.0916	1664 B
Report	1000	10	14,971.7 ns	238.88 ns	13.09 ns	0.9155	15496 B
Matrix	100000	2	869,173.9 ns	78,928.55 ns	4,326.34 ns	-	313 B
MatrixWeighted	100000	2	836,665.3 ns	409,937.06 ns	22,470.03 ns	-	313 B
AccuracyScore	100000	2	165,424.5 ns	4,153.58 ns	227.67 ns	-	-
F1Macro	100000	2	870,502.4 ns	22,318.76 ns	1,223.37 ns	-	473 B
Report	100000	2	871,989.5 ns	52,710.19 ns	2,889.22 ns	-	6545 B
Matrix	100000	10	986,476.4 ns	5,700.15 ns	312.44 ns	-	1249 B
MatrixWeighted	100000	10	937,140.0 ns	21,640.07 ns	1,186.17 ns	-	1249 B
AccuracyScore	100000	10	273,247.0 ns	44,436.45 ns	2,435.71 ns	-	-
F1Macro	100000	10	965,554.2 ns	49,311.61 ns	2,702.94 ns	-	1665 B
Report	100000	10	984,614.0 ns	76,479.34 ns	4,192.09 ns	-	15841 B
Matrix	1000000	2	8,668,171.4 ns	141,239.44 ns	7,741.81 ns	-	324 B
MatrixWeighted	1000000	2	8,457,542.3 ns	362,789.54 ns	19,885.72 ns	-	324 B
AccuracyScore	1000000	2	1,745,944.5 ns	62,563.58 ns	3,429.32 ns	-	-
F1Macro	1000000	2	8,314,837.6 ns	488,311.03 ns	26,765.98 ns	-	484 B
Report	1000000	2	8,536,935.1 ns	540,668.02 ns	29,635.84 ns	-	6572 B
Matrix	1000000	10	9,343,238.2 ns	2,195,702.87 ns	120,353.88 ns	-	1260 B
MatrixWeighted	1000000	10	9,438,244.4 ns	1,202,386.04 ns	65,906.84 ns	-	1260 B
AccuracyScore	1000000	10	2,685,855.8 ns	17,069.90 ns	935.66 ns	-	-
F1Macro	1000000	10	9,781,057.7 ns	2,870,269.92 ns	157,329.18 ns	-	1676 B
Report	1000000	10	9,889,460.7 ns	816,289.51 ns	44,743.58 ns	-	15892 B

Lodestar.Text.Benchmarks.MyersGateBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Band	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Dp	4	73.47 ns	1.069 ns	0.059 ns	1.00	0.00	-	NA
Kernel	4	73.67 ns	7.418 ns	0.407 ns	1.00	0.00	-	NA

Dp	6	108.97 ns	3.137 ns	0.172 ns	1.00	0.00	-	NA
Kernel	6	108.46 ns	8.671 ns	0.475 ns	1.00	0.00	-	NA

Dp	8	158.69 ns	18.662 ns	1.023 ns	1.00	0.01	-	NA
Kernel	8	87.57 ns	0.542 ns	0.030 ns	0.55	0.00	-	NA

Dp	10	204.22 ns	43.249 ns	2.371 ns	1.00	0.01	-	NA
Kernel	10	95.29 ns	21.913 ns	1.201 ns	0.47	0.01	-	NA

Dp	12	250.91 ns	61.649 ns	3.379 ns	1.00	0.02	-	NA
Kernel	12	107.44 ns	7.482 ns	0.410 ns	0.43	0.01	-	NA

Dp	16	410.69 ns	8.257 ns	0.453 ns	1.00	0.00	-	NA
Kernel	16	121.22 ns	4.785 ns	0.262 ns	0.30	0.00	-	NA

Dp	24	887.81 ns	82.390 ns	4.516 ns	1.00	0.01	-	NA
Kernel	24	155.01 ns	2.804 ns	0.154 ns	0.17	0.00	-	NA

Dp	32	1,507.24 ns	460.874 ns	25.262 ns	1.00	0.02	-	NA
Kernel	32	190.91 ns	4.366 ns	0.239 ns	0.13	0.00	-	NA

Dp	48	3,470.11 ns	66.748 ns	3.659 ns	1.00	0.00	-	NA
Kernel	48	261.92 ns	8.048 ns	0.441 ns	0.08	0.00	-	NA

Dp	64	6,311.03 ns	3,411.788 ns	187.012 ns	1.00	0.04	-	NA
Kernel	64	334.03 ns	18.166 ns	0.996 ns	0.05	0.00	-	NA

Dp	96	14,628.02 ns	2,837.071 ns	155.509 ns	1.00	0.01	-	NA
Kernel	96	1,084.25 ns	208.716 ns	11.440 ns	0.07	0.00	-	NA

Lodestar.Text.Benchmarks.PersistenceBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
VocabTxt	4.209 ms	3.9470 ms	0.2163 ms	117.1875	109.3750	39.0625	3.62 MB
TokenizerJsonWordPiece	11.222 ms	1.7109 ms	0.0938 ms	187.5000	171.8750	46.8750	5.72 MB
TokenizerJsonUnigram	12.607 ms	0.2991 ms	0.0164 ms	93.7500	78.1250	31.2500	4.64 MB
SpieceModel	3.804 ms	4.0455 ms	0.2217 ms	109.3750	101.5625	31.2500	3.36 MB
TfidfSave	1.871 ms	0.7248 ms	0.0397 ms	33.2031	27.3438	27.3438	2.09 MB
TfidfLoad	4.431 ms	1.3901 ms	0.0762 ms	85.9375	78.1250	23.4375	2.86 MB
EmbeddingIndexSave	6.710 ms	0.4784 ms	0.0262 ms	476.5625	476.5625	476.5625	39.64 MB
EmbeddingIndexLoad	7.293 ms	0.9727 ms	0.0533 ms	531.2500	500.0000	468.7500	35.35 MB

Lodestar.Text.Benchmarks.StopWordBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Documents	Mean	Error	StdDev	Ratio	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Count	200	7.053 ms	0.6577 ms	0.0360 ms	1.00	500.0000	234.3750	62.5000	7.92 MB	1.00
CountWithStopWords	200	5.969 ms	0.5619 ms	0.0308 ms	0.85	390.6250	187.5000	-	6.34 MB	0.80
Hashing	200	7.004 ms	0.3267 ms	0.0179 ms	0.99	507.8125	171.8750	70.3125	8.08 MB	1.02
HashingWithStopWords	200	6.135 ms	0.2227 ms	0.0122 ms	0.87	406.2500	156.2500	-	6.52 MB	0.82

Count	1000	28.809 ms	2.0211 ms	0.1108 ms	1.00	2718.7500	1031.2500	531.2500	38.64 MB	1.00
CountWithStopWords	1000	22.645 ms	2.2743 ms	0.1247 ms	0.79	1968.7500	781.2500	250.0000	30.99 MB	0.80
Hashing	1000	27.549 ms	2.1014 ms	0.1152 ms	0.96	2562.5000	750.0000	593.7500	39.55 MB	1.02
HashingWithStopWords	1000	23.492 ms	1.2820 ms	0.0703 ms	0.82	2031.2500	625.0000	250.0000	31.83 MB	0.82

Lodestar.Text.Benchmarks.VectorMathBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Dim	Mean	Error	StdDev	Ratio	Allocated	Alloc Ratio
Dot	384	47.82 ns	4.771 ns	0.262 ns	1.00	-	NA
L2Norm	384	48.21 ns	0.518 ns	0.028 ns	1.01	-	NA

Dot	768	94.68 ns	0.717 ns	0.039 ns	1.00	-	NA
L2Norm	768	92.70 ns	0.656 ns	0.036 ns	0.98	-	NA

Dot	1024	123.98 ns	1.124 ns	0.062 ns	1.00	-	NA
L2Norm	1024	122.58 ns	0.802 ns	0.044 ns	0.99	-	NA

Lodestar.Text.Benchmarks.VectorizerBenchmarks-report-github

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

BenchmarkDotNet v0.14.0, Ubuntu 24.04.4 LTS (Noble Numbat)
AMD EPYC 7763, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.400
  [Host]   : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2
  ShortRun : .NET 10.0.11 (10.0.1126.37416), X64 RyuJIT AVX2

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Documents	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Count	200	2.801 ms	1.2783 ms	0.0701 ms	1.00	0.03	93.7500	39.0625	-	1.6 MB	1.00
Tfidf	200	2.876 ms	2.1327 ms	0.1169 ms	1.03	0.04	101.5625	39.0625	-	1.63 MB	1.02
CountBigrams	200	3.652 ms	1.4623 ms	0.0802 ms	1.30	0.04	171.8750	109.3750	-	2.78 MB	1.74
Hashing	200	2.768 ms	0.2241 ms	0.0123 ms	0.99	0.02	93.7500	23.4375	-	1.6 MB	1.00

Count	1000	6.729 ms	0.6616 ms	0.0363 ms	1.00	0.01	492.1875	351.5625	70.3125	7.83 MB	1.00
Tfidf	1000	6.935 ms	0.6365 ms	0.0349 ms	1.03	0.01	492.1875	304.6875	93.7500	7.97 MB	1.02
CountBigrams	1000	10.906 ms	0.4142 ms	0.0227 ms	1.62	0.01	906.2500	375.0000	265.6250	13.42 MB	1.71
Hashing	1000	6.644 ms	0.6207 ms	0.0340 ms	0.99	0.01	492.1875	156.2500	70.3125	7.85 MB	1.00

compare-indel

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

Python: rapidfuzz 3.14.5 (py 3.12.13)
C#:     Lodestar.Text on .NET 10.0.11 (mode Utf16Unit)

length	Python ns/pair	C# ns/pair	speedup (py/C#)
8	117.2	27.1	4.32x C# faster
32	181.3	93.6	1.94x C# faster
128	483.1	966.7	2.00x Py faster
512	4465.7	10777.6	2.41x Py faster

Note: Indel is len(a)+len(b)-2*LCS on both sides, so this compares the subsequence kernels. Lodestar's is a rolling-row dynamic program (#273).

compare-levenshtein

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

Python: rapidfuzz 3.14.5 (py 3.12.13)
C#:     Lodestar.Text on .NET 10.0.11 (mode Utf16Unit)

length	Python ns/pair	C# ns/pair	speedup (py/C#)
8	152.3	19.3	7.90x C# faster
32	281.8	167.2	1.69x C# faster
128	1686.9	1347.9	1.25x C# faster
512	14268.9	14781.5	1.04x Py faster

Note: Python times the realistic per-call loop; rapidfuzz's C core uses the bit-parallel Myers algorithm, so it scales better on long strings.

compare-metrics

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

Python: {'scikit-learn': '1.9.0', 'numpy': '2.5.1'} (py 3.12.13)
C#:     Lodestar on .NET 10.0.11

operation	C# ms	Py ms	wall	C# cpu	Py cpu	cpu
confusion_matrix_n1000_k2	0.009	1.041	113.08x	0.009	1.041	113.08x
accuracy_n1000_k2	0.001	0.516	552.15x	0.001	0.516	552.13x
precision_recall_f1_macro_n1000_k2	0.008	1.737	224.45x	0.008	1.737	224.45x
classification_report_n1000_k2	0.010	7.133	695.36x	0.010	7.132	695.39x
roc_auc_binary_n1000_k2	0.015	1.915	124.25x	0.015	1.915	124.22x
balanced_accuracy_n1000_k2	0.008	1.119	144.73x	0.008	1.119	144.73x
matthews_n1000_k2	0.008	1.982	259.14x	0.008	1.982	259.14x
cohen_kappa_n1000_k2	0.008	1.100	144.42x	0.008	1.100	144.43x
mse_n1000_k2	0.002	0.304	125.05x	0.002	0.304	125.05x
mae_n1000_k2	0.002	0.303	124.51x	0.002	0.303	124.51x
median_ae_n1000_k2	0.006	0.313	55.83x	0.006	0.313	55.82x
r2_n1000_k2	0.002	0.362	146.37x	0.002	0.362	146.37x
confusion_matrix_n1000_k10	0.009	0.985	104.62x	0.009	0.985	104.62x
accuracy_n1000_k10	0.001	0.516	552.64x	0.001	0.516	552.67x
precision_recall_f1_macro_n1000_k10	0.008	1.761	213.09x	0.008	1.761	213.10x
classification_report_n1000_k10	0.015	6.828	462.83x	0.015	6.828	462.81x
roc_auc_ovr_macro_n1000_k10	0.543	9.705	17.88x	0.543	9.705	17.88x
balanced_accuracy_n1000_k10	0.008	1.040	125.05x	0.008	1.040	125.05x
matthews_n1000_k10	0.008	1.994	240.81x	0.008	1.994	240.81x
cohen_kappa_n1000_k10	0.009	1.090	127.05x	0.009	1.090	127.06x
mse_n1000_k10	0.002	0.303	124.94x	0.002	0.303	124.94x
mae_n1000_k10	0.002	0.302	123.88x	0.002	0.302	123.89x
median_ae_n1000_k10	0.006	0.318	56.57x	0.006	0.318	56.56x
r2_n1000_k10	0.002	0.365	147.50x	0.002	0.365	147.50x
confusion_matrix_n100000_k2	0.985	10.660	10.82x	0.985	10.660	10.82x
accuracy_n100000_k2	0.159	3.738	23.51x	0.159	3.737	23.51x
precision_recall_f1_macro_n100000_k2	0.862	12.227	14.18x	0.862	12.226	14.18x
classification_report_n100000_k2	0.860	26.868	31.24x	0.860	26.866	31.23x
roc_auc_binary_n100000_k2	3.538	26.298	7.43x	3.538	26.297	7.43x
balanced_accuracy_n100000_k2	0.863	10.731	12.43x	0.863	10.731	12.43x
matthews_n100000_k2	0.863	21.433	24.82x	0.863	21.431	24.82x
cohen_kappa_n100000_k2	0.855	10.774	12.59x	0.855	10.774	12.59x
mse_n100000_k2	0.479	0.444	0.93x	0.479	0.443	0.93x
mae_n100000_k2	0.473	0.435	0.92x	0.473	0.435	0.92x
median_ae_n100000_k2	0.672	1.785	2.65x	0.698	1.785	2.56x
r2_n100000_k2	0.235	0.662	2.81x	0.235	0.662	2.81x
confusion_matrix_n100000_k10	0.974	10.675	10.96x	0.974	10.672	10.96x
accuracy_n100000_k10	0.257	3.750	14.56x	0.257	3.749	14.56x
precision_recall_f1_macro_n100000_k10	0.985	12.879	13.08x	0.985	12.878	13.08x
classification_report_n100000_k10	0.978	29.403	30.06x	0.978	29.401	30.06x
roc_auc_ovr_macro_n100000_k10	36.595	210.625	5.76x	36.591	210.598	5.76x
balanced_accuracy_n100000_k10	0.980	10.767	10.99x	0.980	10.766	10.99x
matthews_n100000_k10	0.981	22.138	22.57x	0.981	22.135	22.57x
cohen_kappa_n100000_k10	0.990	10.807	10.91x	0.990	10.806	10.91x
mse_n100000_k10	0.476	0.445	0.94x	0.476	0.445	0.94x
mae_n100000_k10	0.479	0.436	0.91x	0.479	0.436	0.91x
median_ae_n100000_k10	0.685	1.786	2.61x	0.729	1.786	2.45x
r2_n100000_k10	0.235	0.660	2.81x	0.235	0.660	2.81x
confusion_matrix_n1000000_k2	8.607	98.325	11.42x	8.606	98.295	11.42x
accuracy_n1000000_k2	1.764	32.718	18.55x	1.764	32.718	18.55x
precision_recall_f1_macro_n1000000_k2	8.607	107.050	12.44x	8.608	107.046	12.44x
classification_report_n1000000_k2	8.571	207.834	24.25x	8.571	207.833	24.25x
roc_auc_binary_n1000000_k2	41.711	285.235	6.84x	41.713	285.212	6.84x
balanced_accuracy_n1000000_k2	8.671	98.533	11.36x	8.670	98.531	11.36x
matthews_n1000000_k2	8.666	198.719	22.93x	8.665	198.707	22.93x
cohen_kappa_n1000000_k2	8.602	98.553	11.46x	8.602	98.547	11.46x
mse_n1000000_k2	2.391	1.937	0.81x	2.391	1.937	0.81x
mae_n1000000_k2	2.367	1.918	0.81x	2.367	1.918	0.81x
median_ae_n1000000_k2	6.255	13.918	2.23x	6.331	13.917	2.20x
r2_n1000000_k2	2.344	3.223	1.38x	2.344	3.223	1.38x
confusion_matrix_n1000000_k10	9.780	98.386	10.06x	9.779	98.384	10.06x
accuracy_n1000000_k10	2.721	32.707	12.02x	2.721	32.705	12.02x
precision_recall_f1_macro_n1000000_k10	9.845	113.111	11.49x	9.844	113.103	11.49x
classification_report_n1000000_k10	9.802	232.322	23.70x	9.801	232.305	23.70x
balanced_accuracy_n1000000_k10	9.861	98.614	10.00x	9.861	98.605	10.00x
matthews_n1000000_k10	9.834	205.474	20.89x	9.833	205.450	20.89x
cohen_kappa_n1000000_k10	9.761	98.626	10.10x	9.760	98.622	10.10x
mse_n1000000_k10	2.395	1.990	0.83x	2.395	1.990	0.83x
mae_n1000000_k10	2.371	1.966	0.83x	2.371	1.966	0.83x
median_ae_n1000000_k10	5.994	13.944	2.33x	6.038	13.943	2.31x
r2_n1000000_k10	2.752	3.418	1.24x	2.752	3.418	1.24x

ratio > 1 means Lodestar is faster. cpu is the merge gate for this branch (docs/guides/performance.md): every operation, every size, must be >= 1x.

BELOW GATE on processor time: mse_n100000_k2 0.93x mae_n100000_k2 0.92x mse_n100000_k10 0.94x mae_n100000_k10 0.91x mse_n1000000_k2 0.81x mae_n1000000_k2 0.81x mse_n1000000_k10 0.83x mae_n1000000_k10 0.83x

compare-persistence

As of 2026-08-19, measured at commit 7946f61772df2cbb0a6c140e557731e594b801fc.

Python: {'tokenizers': '0.23.1', 'sentencepiece': '0.2.2', 'scikit-learn': '1.9.0', 'numpy': '2.5.1'} (py 3.12.13)
C#:     Lodestar on .NET 10.0.11

operation	C# ms	Py ms	wall	C# cpu	Py cpu	cpu
vocab_txt	4.584	9.590	2.09x	4.937	9.590	1.94x
tokenizer_json_wordpiece	10.867	16.095	1.48x	11.335	16.095	1.42x
tokenizer_json_unigram	12.317	34.618	2.81x	13.221	34.615	2.62x
spiece_model	4.660	27.911	5.99x	4.905	27.853	5.68x
tfidf_save	2.052	2.447	1.19x	2.129	2.446	1.15x
tfidf_load	4.274	4.032	0.94x	4.456	4.032	0.90x
embedding_index_save	6.254	4.478	0.72x	6.812	4.478	0.66x
embedding_index_load	6.805	1.190	0.17x	7.629	1.190	0.16x

ratio > 1 means Lodestar is faster. cpu is the honest one: elapsed time hides work .NET does on background GC threads; CPython is single-threaded.

Lodestar

Project

Clone this wiki locally