Only 1 of the Top AI Coding Models on WebDev Arena Isn't Chinese

WebDev AI Leaderboard - Best AI Models for Web Development

New Chat

Leaderboard

Search

Terms of UsePrivacy Policy

Code Arena | WebDev🏆Overall

View overall rankings across AI models on front-end web development tasks, including agentic coding workflows that require multi-step reasoning and tool use.

Jul 1, 2026

422,204 votes

18 labs

Rank by

Lab Rank

Model Score

Rank Spread

1

Anthropic

claude-fable-5 · Proprietary

1653+15/-15

1

11

2

Z.ai

glm-5.2 (max)

1584+12/-12

2

22

3

Bytedance

seed-2.1-pro-preview · Proprietary

1539+13/-13

8

313

4

Alibaba

qwen3.7-max-20260517 · Proprietary

1526+10/-10

11

615

5

Moonshot

kimi-k2.6

1514+8/-8

14

1117

6

Google

gemini-3.5-flash · Proprietary

1510+9/-9

15

1117

7

MiniMax

minimax-m3

1501+10/-10

16

1419

8

OpenAI

gpt-5.5-xhigh (codex-harness) · Proprietary

1501+8/-8

17

1419

9

Xiaomi

mimo-v2.5-pro

1473+8/-8

21

1925

10

DeepSeek

deepseek-v4-pro-thinking

1457+8/-8

26

2231

11

xAI

grok-4.20-beta-0309-reasoning · Proprietary

1383+7/-7

52

4256

12

Tencent

hunyuan-hy3-preview

1362+17/-17

59

5065

13

Poolside

laguna-m.1

1351+13/-13

63

5468

14

Mistral

mistral-medium-3.5

1269+15/-15

76

7482

15

KwaiKAT

KAT-Coder-Pro-V1 · Proprietary

1259+16/-16

77

7683

16

Arcee AI

trinity-large-thinking

1243+19/-19

80

7685

17

IBM

granite-4.1-8b

1200+18/-18

87

8489

18

Inception AI

mercury-2 · Proprietary

1165+23/-23

89

8791

Remove Style Control Leaderboard Plots

Fraction of Model A Wins for All Non-tied A vs. B Battles

Confidence Intervals on Model Strength (via Bootstrapping)

Battle Count for Each Combination of Models (without Ties)

Average Win Rate Against All Other Models (Uniform Sampling and No Ties)