Skip to content

Benchmark

Arena ELO Webdev - Webdev Html

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

Models measured
138
Scores recorded
158
Sources
1
IQ weight
2.06

Score distribution

1,488.6
Mean
1,488.1
Median
154.2
Std dev
1,286.1
10th percentile
1,683
90th percentile
1,078
Min
1,0781,854.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.06 for chat models, measured over 138 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101GPT 5Medium· LMArena data1,426.4
102GPT 5.4· LMArena data1,424.3
103GPT-5.3 CodexCodex Harness· LMArena data1,421.5
104GPT-5.4 miniHigh· LMArena data1,419.8
105DeepSeek-V4-Flash PreviewHigh· LMArena data1,418
106MiniMax M2.5· LMArena data1,413.8
107Kimi K2.5Instant· LMArena data1,411.3
108Qwen3.5 397B A17B· LMArena data1,408.8
109NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,406.6
110GPT 5.1Medium· LMArena data1,406.4
111Inkling FP4· LMArena data1,402.4
112MiniMax M2.7· LMArena data1,401.4
113Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,398.7
114Claude Opus 4.120250805· LMArena data1,398.6
115MiniMax-M2.1 Preview· LMArena data1,398.2
116Inkling Small· LMArena data1,396.8
117Gemini 3 FlashMinimal· LMArena data1,394.6
118Claude Sonnet 4.520250929· LMArena data1,388.5
119Qwen3.5 27B· LMArena data1,385.4
120Gemma 4 31B IT· LMArena data1,379.7
121Solar Pro 4· LMArena data1,371
122Grok 4.20Reasoning, 0309, beta· LMArena data1,371
123DeepSeek V3.2Thinking· LMArena data1,370.2
124Grok 4.3· LMArena data1,366.6
125GPT 5.1· LMArena data1,364.2
126Muse Glimmer· LMArena data1,363.4
127Gemma 4 26B A4B· LMArena data1,361.6
128MiMo V2 FlashNon-Thinking· LMArena data1,353.7
129Qwen3.5 122B-A10B· LMArena data1,352.3
130GLM 4.6· LMArena data1,350.9
131GPT-5.2 Codex· LMArena data1,348.1
132GPT-5.1 Codex· LMArena data1,346.8
133MiMo V2 FlashThinking· LMArena data1,344.2
134Laguna M.1· LMArena data1,334.2
135Kimi K2 Thinking· LMArena data1,325.3
136Claude Haiku 4.520251001· LMArena data1,317.8
137Mistral Medium 3.5· LMArena data1,309.5
138MiniMax M2· LMArena data1,308.1
139Qwen3.5 35B A3B· LMArena data1,306.4
140DeepSeek V3.2· LMArena data1,305.8
141Qwen3.5 Flash· LMArena data1,290.7
142Qwen3 Coder· LMArena data1,286.1
143Laguna XS.2· LMArena data1,285.2
144DeepSeek V3.2 Exp· LMArena data1,280.6
145Gemini 3.1 Flash Lite· LMArena data1,275.5
146KAT-Coder-Pro V1· LMArena data1,266.4
147Grok 4.1 FastReasoning· LMArena data1,253.9
148GPT-5.1 Codex Mini· LMArena data1,252.5
149Mistral Large 3· LMArena data1,240.4
150Gemini 2.5 Pro· LMArena data1,234.9

Showing 101–150 of 158 results.

Data credits: LMArena, CC BY 4.0 (licence)