Skip to content

Model Leaderboard & Specifications

Rankings of AI models evaluated on frontend web generation. Ranked by sample-size adjusted axe-core accessibility score, benchmark activity, and external Artificial Analysis indexes. Price /1M (In·Out) is USD per 1M tokens (input / output). Click a column header to sort.

Rank Model & Provider Top Harness
#1 2026-08-05 8
82/100 axe-core
avg 90/100 · 8 scored runs
ZCode
#2 2026-08-04 8
79/100 axe-core
avg 85/100 · 8 scored runs
ZCode
#3 2026-08-04 8
75/100 axe-core
avg 79/100 · 8 scored runs
Antigravity CLI
#4 2026-08-04 11
74/100 axe-core
avg 77/100 · 11 scored runs
ZCode
#5 2026-08-05 8
74/100 axe-core
avg 77/100 · 8 scored runs
ZCode
#6 2026-08-04 9
73/100 axe-core
avg 75/100 · 9 scored runs
Codex
#7 2026-08-04 8
72/100 axe-core
avg 75/100 · 8 scored runs
ZCode
#8 2026-08-06 8
70/100 axe-core
avg 72/100 · 7 scored runs
ZCode
#9 2026-08-05 9
70/100 axe-core
avg 71/100 · 9 scored runs
ZCode
#10 2026-08-05 8
65/100 axe-core
avg 64/100 · 8 scored runs
ZCode
#11 2026-08-03 3
63/100 axe-core
avg 57/100 · 3 scored runs
Grok CLI
#12 2026-08-05 8
62/100 axe-core
avg 59/100 · 8 scored runs
ZCode
#13 2026-08-05 8
59/100 axe-core
avg 55/100 · 8 scored runs
ZCode
#14 2026-08-06 8
53/100 axe-core
avg 46/100 · 8 scored runs
Command Code
#15 2026-08-04 10
51/100 axe-core
avg 45/100 · 10 scored runs
ZCode
#16 2026-08-04 8
50/100 axe-core
avg 42/100 · 8 scored runs
ZCode