WebDev Arena leaderboard: which LLMs score highest

WebDev Arena (building web apps, crowd-voted (Elo)) currently has scores for 63 models. The top score is 1,820 by Claude Opus 5.5 (max); the median model scores 1,399 and the lowest scores 1,149.

Compare all benchmarks side by side on the live leaderboard →

Among the ten highest scorers, the cheapest listed API price belongs to GLM-5.3 (max) at $0.070 per million input tokens (score 1,614).

What WebDev Arena measures

WebDev Arena asks two anonymous models to build the same web app and lets users vote on the better result; the ratings are Elo-style.

How to read these scores

Scores are Elo-style ratings (higher is better). Only differences between models mean anything, and small gaps may sit inside the source's own margin of error. Where a model is listed at several reasoning efforts, only its highest-effort row is shown. See the methodology for how rows are chosen.

Where the data comes from

Epoch AI Benchmarking Hub (CC BY 4.0). Scores were last captured 2026-10-05; the Captured column gives each row's date.

Related benchmarks

• ALE-Bench
• SciCode
• Terminal-Bench 2.0
• CritPt
• Vending-Bench 2
• SimpleQA Verified