Skip to content

Benchmark

Arena ELO Webdev - Image to Webdev

Coding benchmarks compare models on programming tasks. Higher scores are better.

Models measured
59
Scores recorded
69
Sources
1
IQ weight
2.53

Score distribution

1,540.7
Mean
1,544
Median
100.9
Std dev
1,410.7
10th percentile
1,664
90th percentile
1,284.6
Min
1,284.61,758.4
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.53 for chat models, measured over 59 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Opus 5.5Max· LMArena data1,758.4
2Claude Sonnet 5.5xHigh· LMArena data1,746.4
3Claude Fable 5.1Max· LMArena data1,722.7
4GPT-6 AstraMax· LMArena data1,718.6
5GPT-6.1 SolMax· LMArena data1,694.5
6Qwen4 MaxxHigh· LMArena data1,674.9
7GPT-6.1 TerraMax· LMArena data1,674.2
8Claude Opus 5Max· LMArena data1,664
9GPT-6 SolMax· LMArena data1,659.1
10Qwen3.8 Max0902· LMArena data1,638.6
11Muse Spark 1.3Max· LMArena data1,635.8
12Claude Fable 5High· LMArena data1,620.2
13Hy 4· LMArena data1,618.6
14Qwen3.8 Max· LMArena data1,615.3
15DeepSeek V4.1 Flash 0910Max· LMArena data1,611.8
16Kimi K3.1 ProMax· LMArena data1,608.8
17DeepSeek V4.1 Flash 1017Max· LMArena data1,608.1
18Grok 4.6High· LMArena data1,607.2
19GPT-5.6 SolxHigh, Codex Harness· LMArena data1,606
20Gemini 3.7 FlashHigh· LMArena data1,600.5
21Muse Spark 1.3xHigh· LMArena data1,598.7
22MiMo V2.6 Pro· LMArena data1,595.5
23Gemini 3.8 FlashHigh· LMArena data1,592.5
24MiMo V2.6 Flash· LMArena data1,592.1
25Grok 4.7xHigh· LMArena data1,591.5
26Kimi K3Max· LMArena data1,586.4
27GLM 5.3 Flash· LMArena data1,581.4
28Claude Opus 4.7High· LMArena data1,579.4
29Qwen 3.8 27B· LMArena data1,574.9
30GPT-6 LunaMax· LMArena data1,567.5
31Claude Opus 4.7· LMArena data1,562
32Grok 4.5· LMArena data1,561.5
33Step 5 PreviewHigh· LMArena data1,550.2
34Gemini 3.6 FlashHigh· LMArena data1,544.4
35Claude Opus 4.6High· LMArena data1,544
36GPT-5.6 TerraxHigh, Codex Harness· LMArena data1,539.5
37Claude Sonnet 5High· LMArena data1,538.9
38Muse Spark 1.1· LMArena data1,538.8
39Claude Opus 4.6· LMArena data1,537.7
40GPT 5.5xHigh, Codex Harness· LMArena data1,531.4
41Claude Sonnet 4.6· LMArena data1,529.3
42Kimi K2.6· LMArena data1,522.1
43GPT 5.5High, Codex Harness· LMArena data1,519.1
44GPT-5.6 LunaxHigh, Codex Harness· LMArena data1,517.5
45Kimi K2.7 Code· LMArena data1,511.4
46Gemini 3.5 FlashMedium· LMArena data1,505
47Seed2.1 Pro Preview· LMArena data1,504.8
48GPT 5.5Codex Harness· LMArena data1,503.5
49Gemini 3.5 FlashHigh· LMArena data1,500
50Gemini 3.1 Pro Preview· LMArena data1,479.9

Showing 1–50 of 69 results.

Data credits: LMArena, CC BY 4.0 (licence)