ENفا
Get API key
MODELSArena’s interaction judge tests what agents actually finish▼ 22%DeepSeek V4.1 Flash API price#1Claude Fable 5.2 · 30.48% win shareFUNDINGAnthropic targets $64.4B Nasdaq IPO as revenue surgesTOOLSCursor tops the tools directory this weekMODELSGoogle cuts Gemini 4 Carbon prices before Gemini 2.5 sunsetPOLICYFive frontier AI labs face coordinated SF safety lawsuits
Vigas Arena · updated just now

The best WebDev models,
ranked by real use.

Evaluates full-stack web UI quality, responsive CSS layout fidelity, accessibility, and interactive component architecture.

378,706,203sessions rankedHow we rank →

Rising this week · WebDev

Claude Fable 5.2up 3.79 pts · #1 in WebDevClaude Opus 5.5up 1.86 pts · #2 in WebDevGLM-5.3up 1.41 pts · #9 in WebDev

The podium · WebDev

Win share · this week
  1. Rank 1: Claude Fable 5.2Anthropic · 1838 score31.80%
  2. Rank 2: Claude Opus 5.5Anthropic · 1815 score23.10%
  3. Rank 3: Claude Sonnet 5.5Anthropic · 1772 score18.40%

Full ranking

WebDev leaderboard: win share, score, cost per task and votes
#ModelVotes7 days
1Claude Fable 5.2 (xHigh)Anthropic31.80%1838 ±21$7.40142,000
2Claude Opus 5.5 (Max)Anthropic23.10%1815 ±12$5.71134,733
3Claude Sonnet 5.5 (xHigh)Anthropic18.40%1772 ±10$1.95128,466
4Gemini 4 Carbon (High)Google17.95%1768 ±15$1.42121,199
5GPT-6 Astra (Max)OpenAI17.20%1764 ±9$4.10115,932
6GPT-6.1 Sol (Max)OpenAI15.10%1743 ±10$1.03109,665
7Claude Fable 5.1 (Max)Anthropic14.40%1732 ±14$1.25102,398
8Kimi K3.1 ProMoonshot11.30%1698 ±16$0.3194,131
9GLM-5.3Z.ai10.20%1685 ±18$0.4287,864
10DeepSeek V4.1 FlashDeepSeek8.15%1642 ±14$0.1291,597
#131.80%
Score 1838 ±21$7.40/task142,000 votes
#223.10%
Score 1815 ±12$5.71/task134,733 votes
#318.40%
Score 1772 ±10$1.95/task128,466 votes
Score 1768 ±15$1.42/task121,199 votes
#517.20%
Score 1764 ±9$4.10/task115,932 votes
#615.10%
Score 1743 ±10$1.03/task109,665 votes
#714.40%
Score 1732 ±14$1.25/task102,398 votes
#811.30%
Score 1698 ±16$0.31/task94,131 votes
#910.20%
Score 1685 ±18$0.42/task87,864 votes
#108.15%
Score 1642 ±14$0.12/task91,597 votes

Pareto frontier

score vs $/task

Green models give the best score for their price. Tap any dot to inspect value vs. performance.

Claude Fable 5.2Pareto optimal
Score 1838 · 31.80% win · $7.40/task
API pricing

Live sessions

LIVE

No sessions in the last 24 hours yet. Send a request through the Vigas API.

Connect via API — start in 30s
§ Methodology

How the rankings are made.

Real sessions, not benchmarks

Win share counts how often a model finishes real tasks that users accept — across real API workloads, coding agents, and blind Arena tests.

Blind pairwise votes

In Arena, people compare two hidden models and vote. Scores use a Bradley–Terry model with 95% confidence intervals.

Cost is part of the score

Every ranking shows dollars per completed task, so you can see who is best for the money — not just best.

§ WebDev · Questions & explore

About the webdev board.

Web & UI buildersTools newsAPI pricesAPI quickstart

What is the #1 AI model on the WebDev leaderboard right now?

Claude Fable 5.2 (Anthropic) leads webdev with a 31.80% real-work win share and a Bradley-Terry score of 1838, ahead of #2 Claude Opus 5.5 (23.10%).

Which top-10 model offers the lowest cost per task for webdev?

DeepSeek V4.1 Flash balances #10 performance (score 1642) at $0.12 per completed task on Vigas.

How are Vigas leaderboard scores and win shares calculated?

Vigas combines real session outcomes across API workloads and agent harnesses with blind pairwise Arena votes using a Bradley-Terry statistical model with 95% confidence intervals.

Can I call the top webdev models through one API?

Yes. Point any OpenAI-compatible client at https://api.vigas.ai/v1 with your Vigas API key or set model to vigas/auto to route across providers automatically.