Axis leaderboard / Speed / weight 7 of 100

Fastest AI app builders

Median wall clock seconds per vibeOps prompt, with the p10 to p90 spread. Wall clock time from submitting a prompt to a deployed, reachable artifact.

Ranked by speed

Ordered by median seconds per prompt, fastest first.

Fastest AI app builders
#ToolSpeedp10 / median / p90Spread ratioComposite
100.0
19.8/31.4/54.5
1.74x72.17
79.7
22.5/39.4/75.7
1.92x67.27
79.1
24.2/39.7/64.4
1.62x74.04
58.8
28.7/53.4/82.7
1.55x78.65
52.5
36.1/59.8/110.7
1.85x66.89
48.2
44.9/65.2/104.1
1.60x74.69
47.9
39.9/65.5/100.7
1.54x80.30
43.9
44.9/71.6/110.0
1.54x75.92
43.4
49.1/72.4/140.4
1.94x76.31
39.9
45.0/78.7/123.1
1.56x71.91
39.2
46.3/80.0/143.0
1.79x56.55
38.2
51.8/82.1/136.0
1.66x71.94
37.5
54.4/83.8/149.9
1.79x64.77
36.9
52.6/85.2/133.1
1.56x77.87
36.1
49.4/87.0/137.1
1.58x55.45
35.4
51.5/88.7/142.4
1.61x70.50
34.8
53.4/90.3/168.5
1.87x59.33
34.5
46.1/91.1/174.2
1.91x66.17
33.4
58.8/94.1/161.4
1.72x70.54
33.4
55.2/94.1/146.8
1.56x67.16
33.0
47.4/95.1/201.7
2.12x71.51
32.7
53.5/96.0/186.6
1.94x63.18
32.6
54.3/96.4/202.9
2.10x61.04
32.5
38.2/96.7/167.9
1.74x62.25
31.6
66.3/99.3/189.8
1.91x67.13
31.3
65.8/100.3/161.3
1.61x78.06
29.8
61.1/105.2/186.8
1.78x55.67
29.5
59.2/106.3/167.6
1.58x87.96
29.3
53.3/107.2/217.0
2.02x60.16
29.0
65.1/108.4/176.3
1.63x67.30
27.7
60.2/113.4/193.1
1.70x58.78
27.6
59.6/113.8/231.5
2.03x58.17
26.0
69.6/120.9/243.9
2.02x56.79
23.1
92.8/135.7/188.8
1.39x73.42
22.9
72.2/137.1/249.0
1.82x55.27
21.7
81.7/144.6/309.3
2.14x66.79
21.7
81.4/144.8/269.1
1.86x65.18
21.4
84.4/146.5/292.7
2.00x68.69
20.8
76.0/150.6/295.5
1.96x60.10
19.4
84.8/161.7/350.8
2.17x71.52
8.9
230.2/353.6/651.4
1.84x66.36

How the speed axis is measured

Timed by the harness, not by the product interface, from request submission to a 200 response on the deployed URL. We publish p10, median and p90 across all recorded executions. The subscore is the fastest median in the cohort divided by the product median, expressed as a percentage.

This axis carries 7 of the 100 index points. Each product completed 5 runs of 9 prompts in this cycle, and the underlying per run data behind every figure on this page is downloadable at /api/runs.json under a CC BY 4.0 licence.

A high score on one axis is not a recommendation. Read it against the composite column on the right and against the axes that matter for the work you have in front of you. Reliability at 71.1% for the leader of this cohort is the context for everything else on the page.

Other axis leaderboards