Skip to content

Benchmark

Arena ELO Webdev - Webdev

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

Models measured
139
Scores recorded
159
Sources
1
IQ weight
1.96

Score distribution

1,484.4
Mean
1,484.4
Median
157.2
Std dev
1,263.4
10th percentile
1,683
90th percentile
1,072
Min
1,0721,823.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.96 for chat models, measured over 139 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51GPT-6 LunaMax· LMArena data1,578.3
52Mistral Large 4· LMArena data1,573.2
53Step 5 PreviewHigh· LMArena data1,570.1
54Claude Opus 4.7High· LMArena data1,557.4
55Claude Opus 4.7· LMArena data1,555.5
56Claude Opus 4.8High· LMArena data1,555.5
57Grok 4.5· LMArena data1,552.4
58Claude Opus 4.6High· LMArena data1,546.4
59Muse Spark 1.1· LMArena data1,541.3
60Claude Sonnet 5High· LMArena data1,539.4
61Gemini 3.6 FlashHigh· LMArena data1,537.8
62Claude Opus 4.6· LMArena data1,536.7
63Claude Opus 4.8· LMArena data1,536.5
64Muse Spark 1.2xHigh· LMArena data1,533.3
65Muse Spark OpenMax· LMArena data1,527.3
66Claude Sonnet 4.6· LMArena data1,521.9
67GPT-5.6 TerraxHigh, Codex Harness· LMArena data1,520.4
68GPT-5.6 LunaxHigh, Codex Harness· LMArena data1,519.1
69Seed2.1 Pro Preview· LMArena data1,518.3
70Qwen3.7 Max20260517· LMArena data1,515.7
71GPT 5.5xHigh, Codex Harness· LMArena data1,512.4
72Kimi K2.6· LMArena data1,509.1
73GLM 5.1· LMArena data1,508.1
74Hy3 (Deep Infra)· LMArena data1,505.8
75Gemini 3.5 FlashHigh· LMArena data1,499.8
76Claude Opus 4.5High, 32K, 20251101· LMArena data1,493.6
77Gemini 3.5 FlashMedium· LMArena data1,490.1
78GPT 5.5High, Codex Harness· LMArena data1,487.3
79Spark-X2.6· LMArena data1,485.7
80MiniMax M3· LMArena data1,484.4
81Qwen3.6 Max Preview· LMArena data1,481.3
82MiMo V2.5 Pro· LMArena data1,479.3
83Kimi K2.7 Code· LMArena data1,474.9
84Claude Opus 4.520251101· LMArena data1,468.9
85DeepSeek-V4-Pro PreviewHigh· LMArena data1,463.8
86Qwen3.6 Plus· LMArena data1,461.9
87GPT 5.4High, Codex Harness· LMArena data1,461.5
88GPT 5.5Codex Harness· LMArena data1,455.7
89DeepSeek V4 Pro· LMArena data1,448.4
90Gemini 3.1 Pro Preview· LMArena data1,446.3
91GPT 5.4Medium, Codex Harness· LMArena data1,441.7
92Gemini 3 Pro Preview· LMArena data1,441.1
93Gemini 3 Flash· LMArena data1,440.1
94MiMo V2.5· LMArena data1,437.9
95Beam 1.0 501B A23B· LMArena data1,436.8
96Kimi K2.5Thinking· LMArena data1,436.4
97GLM 4.7· LMArena data1,435.9
98GLM 5· LMArena data1,434.9
99MiMo V2 Pro· LMArena data1,434.8
100Gemini 3.5 Flash Lite· LMArena data1,433.4

Showing 51–100 of 159 results.

Data credits: LMArena, CC BY 4.0 (licence)