LLMTier
Menu

Leaderboard / full-stack app builders / cycle 04

Full-stack AI app builder leaderboard 2026

Every measurement we hold on 20 full-stack app builders, in one grid. 5 runs of the 9 prompt vibeOps specification per product. You give this class a written specification and it returns a hosted, running application; the coding agent board at /leaderboard/coding-agents is a separate measurement with its own weights. Scroll sideways for the full axis set. The first column stays put.
Last run: 15 Aug 2026 09:11 UTCCadence: roughly every seven days, next date announced once it is scheduled

Full-stack app builders

20

This board only, see /leaderboard/coding-agents for the other

Prompt executions

900

Published individually at /api/runs.json

Index leader

Totalum

Composite 89.48

Fastest median turn

65.2 s per prompt

v0 - 9.8 min across the full nine prompt specification, derived

Full measurement grid

Index, turn latency percentiles, outcome split and all ten subscores. Delta is not shown for cycle 04, see below.

CSV | JSON
Full Full-stack app builders leaderboard with per axis subscores and raw measurements
#ToolLab indexReader indexDeltaTurn latency p10/med/p90p90 / medianp90 / p10Full spec (derived)Reliability %Runs pass/part/failAgentScal.SEOAPIIntegr.DesignValueCode own.HTML no JSLCP msCLSEntry price
1Totalum logoTotalum
89.48
72.29withheld
59.2/106.3/167.6
1.58x2.83x15.9 min93.3%
42/2/1
94.092.095.097.088.086.069.096.041.0 KB1,4800.020EUR 29 / month
2Firebase Studio logoFirebase StudioSunsets Mar 2027
80.92
74.74withheld
65.8/100.3/161.3
1.61x2.45x15.0 min86.7%
39/5/1
79.087.073.079.090.072.092.084.022.0 KB1,9800.050No cost
3Lovable logoLovable
80.54
82.79withheld
52.6/85.2/133.1
1.56x2.53x12.8 min82.2%
37/6/2
89.078.075.066.086.092.072.079.018.0 KB2,1400.060USD 25 / month, billed monthly
4v0 logov0
78.31
80.68withheld
44.9/65.2/104.1
1.60x2.32x9.8 min84.4%
38/6/1
80.072.076.060.066.094.068.086.027.0 KB1,6200.040USD 30 / user / month
5Replit logoReplit
75.47
67.02withheld
92.8/135.7/188.8
1.39x2.03x20.4 min86.7%
39/4/2
83.080.062.071.077.073.072.082.09.0 KB2,8800.120USD 20 / month, billed monthly
6Bolt.new logoBolt.new
74.68
77.49withheld
51.8/82.1/136.0
1.66x2.63x12.3 min80.0%
36/8/1
81.070.066.058.072.085.071.088.012.0 KB2,4600.090USD 25 / month, billed monthly
7Base44 logoBase44
73.05
72.48withheld
58.8/94.1/161.4
1.72x2.74x14.1 min84.4%
38/7/0
75.068.064.066.075.080.076.055.014.0 KB2,3200.080USD 16 / month, billed annually
8Anything logoAnything
69.97
71.63withheld
55.2/94.1/146.8
1.56x2.66x14.1 min77.8%
35/8/2
71.062.068.057.066.083.076.070.018.0 KB1,9600.050USD 19 / month
9Softgen logoSoftgen
69.48
1 votingwithheld
65.1/108.4/176.3
1.63x2.71x16.3 min73.3%
33/7/5
74.068.066.058.070.078.070.074.017.0 KB2,1800.060USD 25 / month
10Emergent logoEmergent
68.68
2 votingwithheld
81.7/144.6/309.3
2.14x3.79x21.7 min80.0%
36/3/6
79.066.058.061.063.071.074.072.08.0 KB3,1200.140USD 20 / month, billed monthly
11Floot logoFloot
68.66
1 votingwithheld
46.1/91.1/174.2
1.91x3.78x13.7 min68.9%
31/8/6
72.066.070.055.064.084.070.062.019.0 KB2,0200.050USD 25 / month
12Dyad logoDyad
67.49
72.53withheld
54.4/83.8/149.9
1.79x2.76x12.6 min55.6%
25/10/10
68.064.072.060.062.072.090.098.020.0 KB1,8800.040No licence fee, bring your own model key
13Bind AI logoBind AI
65.59
-withheld
53.5/96.0/186.6
1.94x3.49x14.4 min64.4%
29/7/9
69.060.058.062.068.070.075.080.013.0 KB2,2400.070USD 18 / month
14Rocket.new logoRocket.new
62.12
1 votingwithheld
53.3/107.2/217.0
2.02x4.07x16.1 min60.0%
27/7/11
70.061.057.052.059.075.066.058.07.0 KB3,2600.160USD 25 / month
15Caffeine logoCaffeine
62.08
-withheld
76.0/150.6/295.5
1.96x3.89x22.6 min66.7%
30/12/3
70.074.048.052.046.068.084.058.06.0 KB2,8400.110USD 5 / month
16Zite logoZite
61.06
1 votingwithheld
60.2/113.4/193.1
1.70x3.21x17.0 min60.0%
27/8/10
68.058.060.047.055.077.076.048.010.0 KB2,7400.110USD 25 / month, billed monthly
17bolt.diy logobolt.diy
59.85
2 votingwithheld
59.6/113.8/231.5
2.03x3.88x17.1 min55.6%
25/10/10
60.052.062.048.054.074.082.097.015.0 KB2,3200.080No licence fee, bring your own model key
18YouWare logoYouWare
58.53
2 votingwithheld
46.3/80.0/143.0
1.79x3.09x12.0 min48.9%
22/9/14
63.054.061.044.052.082.0n/a52.014.0 KB2,1200.070Not published by the vendor
19Trickle logoTrickle
58.11
2 votingwithheld
49.4/87.0/137.1
1.58x2.78x13.1 min51.1%
23/11/11
62.052.063.041.049.079.074.045.015.0 KB2,2600.070USD 17 / month
20CatDoes logoCatDoes
56.99
-withheld
72.2/137.1/249.0
1.82x3.45x20.6 min48.9%
22/10/13
66.058.042.048.060.080.074.066.08.0 KB2,4600.090USD 17 / month

Rank deltas are not shown for cycle 04 because the roster was re-based from a single merged ranking into two class boards, so this cycle's ranks and the previous cycle's are not on the same scale. Deltas resume in cycle 05.

Turn latency columns are median wall clock seconds for one prompt, not for a whole build, measured by the harness from submission to the first 200 response on the deployed URL. The full spec column derives what nine of those medians would add up to and is not separately measured. Runs are split pass, partial and fail across the 45 recorded executions per product. Reliability is the strict pass share. HTML no JS is real text bytes with scripts disabled.

Read a single axis instead

The composite compresses ten different questions into one number. Each axis page shows both boards.

Turn latency

Median wall clock seconds for one prompt, not for a whole build. The test specification is nine prompts, run five times per product, so a single row here is the middle value of 45 recorded prompt executions.

Reliability

Strict pass rate across 900 recorded prompt executions.

SEO and GEO

Server rendered HTML bytes with JavaScript disabled, LCP, CLS and structured data output.

API and MCP

REST surface, webhook delivery, background jobs and Model Context Protocol support.

Design output

Blind scored visual quality of the generated admin panel and marketing surface.

Value

Index points delivered per euro of entry price. The two boards price different things, so the two value scores are not comparable across boards.

Notes on reading this table

Products within about one and a half index points of each other should be read as tied. Five runs per product gives a usable median and a rough sense of spread, not a tight confidence interval, and the honest thing to do with a small sample is to say so rather than to present a false ordering.

This table holds only the full-stack app builders: products where you give a written specification and get back a hosted, running application at a deployed URL. Coding agents, which work inside a repository, terminal or IDE you already control and return a diff rather than a deployment, are a separate board with their own nine axis weights at /leaderboard/coding-agents. The two are never merged into one ordering on this site: an index score from one board is not comparable to an index score from the other.

Reliability at 93.3% at the top of the table means roughly one execution in eight still needed intervention or failed outright, for the best product in the cohort. That is the state of the field in August 2026, and it is the single most useful thing to know before planning a delivery date around one of these tools.