JAM

Conformance Performance

Important Note

This leaderboard highlights performance differences between JAM implementations. All implementations are works in progress and none are fully conformant yet. The rankings serve to track relative performance improvements over time.

Performance Comparison

All implementations relative to polkajam(aggregate weighted scores - see methodology below)

Apr 11, 8:24 AM

1062052

Jamzilla

baseline2.19ms

JavaJAM

Java

1.5x slower3.28ms

jamzilla-int

1.3x slower2.88ms

TurboJam

C++

3.1x slower6.73ms

Vinwolf

Rust

4.7x slower10.35ms

Typeberry

TypeScript

5.4x slower11.84ms

JAM Forge

Scala

7.3x slower16.04ms

jotl

10.4x slower22.81ms

Boka

Swift

13.3x slower29.18ms

PyJAMaz

Python

14.7x slower32.22ms

jampy-recompiler

19.1x slower41.82ms

New JAMneration

17.5x slower38.24ms

GrayMatter

Elixir

17.9x slower39.11ms

Jam4s

Scala

21.5x slower47.18ms

JamPy

Python

22.2x slower48.64ms

PBnJAM

TypeScript

34.5x slower75.56ms

Linear scale • Lower is better

Percentiles:

P50

P90

P99

<1.2x

<2x

<10x

>50x

Performance Rankings

Baseline: polkajam(Score: )

Rank	Team	Language	Score	P50 (ms)	P90 (ms)	Relative Performance
1	Jamzilla	Go	3.0	1.54	3.40	baseline
2	JavaJAM	Java	3.9	3.01	5.24	1.5x slower
3	jamzilla-int		4.3	1.79	5.19	1.3x slower
4	TurboJam	C++	7.8	5.59	10.71	3.1x slower
5	Vinwolf	Rust	13.0	10.29	17.65	4.7x slower
6	Typeberry	TypeScript	15.3	10.03	21.28	5.4x slower
7	JAM Forge	Scala	21.5	14.55	22.97	7.3x slower
8	jotl		32.3	17.16	39.09	10.4x slower
9	Boka	Swift	39.0	25.31	48.12	13.3x slower
10	PyJAMaz	Python	42.5	35.21	49.59	14.7x slower
11	jampy-recompiler		55.3	31.99	79.83	19.1x slower
12	New JAMneration	Go	55.5	26.81	61.02	17.5x slower
13	GrayMatter	Elixir	58.9	26.40	62.83	17.9x slower
14	Jam4s	Scala	64.3	44.21	82.60	21.5x slower
15	JamPy	Python	66.1	36.78	95.36	22.2x slower
16	PBnJAM	TypeScript	91.0	61.60	114.04	34.5x slower

Audit Time Calculator

Baseline Audit Time (days)

Time required for polkajam to complete audit

1Jamzilla

3.0d

2JavaJAM

4.5d

3jamzilla-int

3.9d

4TurboJam

9.2d

5Vinwolf

14.2d

6Typeberry

16.2d

7JAM Forge

22.0d

8jotl

31.2d

9Boka

39.9d

10PyJAMaz

44.1d

11jampy-recompiler

57.3d

12New JAMneration

52.4d

13GrayMatter

53.6d

14Jam4s

64.6d

15JamPy

66.6d

16PBnJAM

103.4d

Note: These calculations show the real-world impact of performance differences on audit requirements.

Scoring Methodology

Weighted scoring system that considers full performance distribution. Our scoring system prioritizes consistent, predictable performance by weighing multiple statistical metrics:

Median (P50)

35%

Typical performance

90th Percentile

25%

Consistency

Mean

20%

Average

99th Percentile

10%

Worst case

Consistency

10%

Lower variance

How it works:

1. Performance measurements are based on the public W3F test vector traces
2. For each benchmark, we calculate a weighted score using the metrics above
3. We use geometric mean across all benchmarks to aggregate metrics
4. Teams are ranked by their final weighted score (lower is better)
5. Polkajam (interpreted) serves as the baseline (1.0x) for relative comparisons

Note: Only teams with data for all four benchmarks (Safrole, Fallback, Storage, Storage Light) are included in the overview. Zero values are excluded from calculations as they likely represent measurement errors.

Performance data updated regularly. Version: 0.7.2| Last updated: Apr 11, 2026, 8:24 AM| Source data from: Apr 11, 2026

Testing protocol conformance at scale. Learn more at jam-conformance | Commit 1062052 | View all clients | Download raw data