WebDev Arena Leaderboard

WebDev Arena is a real-time AI coding competition where models go head-to-head in web development challenges, developed by LMArena

Leaderboard

Arena Score

1433.16

License

Proprietary

95% CI

+13.78 / -16.08

Votes

2,464

Arena Score

1408.84

License

MIT

95% CI

+16.75 / -15.04

Votes

1,708

Arena Score

1405.51

License

Proprietary

95% CI

+12.56 / -12.44

Votes

3,622

Arena Score

1381.76

License

Proprietary

95% CI

+17.04 / -18.98

Votes

2,636

Arena Score

1357.03

License

Proprietary

95% CI

+10.78 / -9.24

Votes

7,481

Arena Score

1304.86

License

Proprietary

95% CI

+11.06 / -12.70

Votes

3,084

Arena Score

1256.52

License

Proprietary

95% CI

+6.94 / -8.34

Votes

5,770

Arena Score

1237.67

License

Proprietary

95% CI

+4.49 / -4.79

Votes

26,338

Arena Score

1206.74

License

MIT

95% CI

+20.70 / -18.30

Votes

1,097

DeepSeek-R1

DeepSeek

#9

Arena Score

1198.32

License

MIT

95% CI

+8.45 / -10.08

Votes

3,769

Arena Score

1189.94

License

Proprietary

95% CI

+8.98 / -7.83

Votes

4,954

Arena Score

1189.12

License

Proprietary

95% CI

+10.12 / -11.23

Votes

3,886

Arena Score

1186.82

License

Apache 2.0

95% CI

+13.29 / -11.79

Votes

2,953

Arena Score

1174.50

License

Proprietary

95% CI

+12.92 / -14.76

Votes

2,542

Arena Score

1156.94

License

Proprietary

95% CI

+9.17 / -8.69

Votes

5,242

Arena Score

1142.85

License

Proprietary

95% CI

+7.48 / -8.48

Votes

6,284

Arena Score

1136.37

License

Proprietary

95% CI

+11.69 / -8.97

Votes

2,984

Arena Score

1133.19

License

Proprietary

95% CI

+4.74 / -4.09

Votes

22,128

Arena Score

1102.28

License

Proprietary

95% CI

+9.36 / -11.83

Votes

3,390

Arena Score

1091.78

License

Proprietary

95% CI

+8.97 / -8.66

Votes

6,391

Arena Score

1088.91

License

Proprietary

95% CI

+7.04 / -7.32

Votes

11,936

#22

Arena Score

1044.96

License

Proprietary

95% CI

+7.35 / -5.17

Votes

9,271

Arena Score

1041.86

License

Proprietary

95% CI

+6.20 / -6.11

Votes

13,828

Arena Score

1039.97

License

Proprietary

95% CI

+7.22 / -5.39

Votes

10,533

Arena Score

1029.61

License

Proprietary

95% CI

+21.11 / -24.34

Votes

1,064

Arena Score

1027.32

License

Llama 4

95% CI

+7.91 / -8.54

Votes

5,483

Arena Score

980.35

License

Proprietary

95% CI

+6.50 / -5.73

Votes

14,485

#27

Arena Score

975.06

License

Proprietary

95% CI

+6.06 / -6.16

Votes

11,110

Arena Score

964.00

License

Proprietary

95% CI

+5.12 / -4.41

Votes

18,637

DeepSeek-V3

DeepSeek

#29

Arena Score

959.85

License

DeepSeek

95% CI

+8.29 / -8.00

Votes

7,717

Arena Score

902.25

License

Apache 2.0

95% CI

+6.53 / -5.43

Votes

16,252

Arena Score

900.24

License

Llama 4

95% CI

+29.66 / -29.17

Votes

692

Arena Score

892.55

License

Proprietary

95% CI

+5.82 / -7.05

Votes

15,201

Arena Score

809.75

License

Llama 3.1

95% CI

+16.38 / -19.01

Votes

1,117

More Statistics for WebDev Arena (Overall)

Confidence Interval for Model Strength

Figure 1

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

Figure 2

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

Figure 3

Battle Count for Each Combination of Models (without Ties)

Figure 4