Skip to content

Benchmark

Arena ELO Webdev - Webdev React

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

Models measured
124
Scores recorded
143
Sources
1
IQ weight
2.03

Score distribution

1,506.4
Mean
1,505
Median
159.1
Std dev
1,312.4
10th percentile
1,713.4
90th percentile
1,142.7
Min
1,142.71,885
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.03 for chat models, measured over 124 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Inkling FP4· LMArena data1,404.3
102Inkling Small· LMArena data1,396.5
103Gemini 3 Pro Preview· LMArena data1,396.3
104NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,394.4
105GPT-5.3 CodexCodex Harness· LMArena data1,393.7
106Kimi K2.5Instant· LMArena data1,391.3
107GPT 5.4· LMArena data1,387.7
108Qwen3.5 397B A17B· LMArena data1,387
109MiniMax M2.7· LMArena data1,386.9
110GPT-5.4 miniHigh· LMArena data1,386.4
111Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,385.7
112Claude Sonnet 4.520250929· LMArena data1,382.5
113Gemini 3 FlashMinimal· LMArena data1,373.6
114MiniMax M2.5· LMArena data1,372.8
115Grok 4.20Reasoning, 0309, beta· LMArena data1,368.5
116MiniMax-M2.1 Preview· LMArena data1,364.6
117Gemma 4 31B IT· LMArena data1,362
118Solar Pro 4· LMArena data1,360.1
119Grok 4.3· LMArena data1,352.2
120Qwen3.5 122B-A10B· LMArena data1,351.3
121DeepSeek V3.2Thinking· LMArena data1,349.4
122Gemma 4 26B A4B· LMArena data1,348.2
123Muse Glimmer· LMArena data1,345.8
124Qwen3.5 27B· LMArena data1,344.1
125Laguna M.1· LMArena data1,338.7
126DeepSeek V3.2· LMArena data1,332.5
127GPT-5.2 Codex· LMArena data1,328.8
128Claude Haiku 4.520251001· LMArena data1,325.4
129Kimi K2 Thinking· LMArena data1,312.4
130MiMo V2 FlashNon-Thinking· LMArena data1,312.2
131GPT 5.1· LMArena data1,306.6
132Laguna XS.2· LMArena data1,295.4
133Qwen3 Coder· LMArena data1,257.2
134Mistral Medium 3.5· LMArena data1,252.3
135Gemini 3.1 Flash Lite· LMArena data1,243.6
136Qwen3.5 35B A3B· LMArena data1,235.8
137Trinity Large Thinking· LMArena data1,230.4
138Qwen3.5 Flash· LMArena data1,228.4
139MiMo V2 FlashThinking· LMArena data1,226.5
140Grok 4.1 FastReasoning· LMArena data1,220.2
141Granite 4.1 8B· LMArena data1,180.6
142Mercury 2· LMArena data1,159.2
143Devstral 2 123b Instruct 2512· LMArena data1,142.7

Showing 101–143 of 143 results.

Data credits: LMArena, CC BY 4.0 (licence)