Skip to content

Benchmark

Arena ELO - Coding

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

Models measured
361
Scores recorded
418
Sources
1
IQ weight
1.28

Score distribution

1,325
Mean
1,374.5
Median
162.1
Std dev
1,081.6
10th percentile
1,488
90th percentile
682.8
Min
682.81,564.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.28 for chat models, measured over 361 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51Claude Sonnet 4.520250929· LMArena data1,484.9
52Kimi K2.5Instant· LMArena data1,484.1
53Kimi K3.1 ProMax· LMArena data1,483.9
54nova-experimental-chat26-02-10· LMArena data1,483.2
55Claude Sonnet 5High· LMArena data1,482.8
56Gemini 3.1 Pro Preview· LMArena data1,482.7
57GPT 5.5· LMArena data1,482.5
58GPT-6.1 TerraMax· LMArena data1,481.9
59Muse Spark· LMArena data1,481.6
60Step 5 Preview· LMArena data1,481.2
61Gemini 3 Pro Preview· LMArena data1,481.2
62Qwen 3.8 27B· LMArena data1,481.1
63Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,478.9
64GPT 5.4· LMArena data1,478.6
65MiMo V2 Pro· LMArena data1,475.9
66Beam 1.0 501B A23B· LMArena data1,474.6
67Grok 4.5· LMArena data1,474.3
68Kimi K2.5Thinking· LMArena data1,474
69Claude Opus 4.120250805· LMArena data1,472.4
70Qwen3.7 Plus· LMArena data1,472.3
71Doubao Seed 2.0 Pro· LMArena data1,471.9
72LongCat-Flash-Chat Exp2602· LMArena data1,470.5
73Qwen3.6 Max Preview· LMArena data1,470.3
74MiMo V2.5· LMArena data1,469.3
75DeepSeek V4 Pro· LMArena data1,469.2
76Nemotron 3 UltraNVFP4· LMArena data1,467.7
77Hy3 (Deep Infra)· LMArena data1,467.7
78MiniMax M3· LMArena data1,467.3
79Qwen3.6 Plus· LMArena data1,467.3
80GPT-5.6 LunaxHigh· LMArena data1,467.3
81Inkling FP4· LMArena data1,466.8
82DeepSeek V4 ProHigh, 20260813· LMArena data1,466.8
83LongCat Flash Chat· LMArena data1,466.4
84MiMo V2 Omni· LMArena data1,465.9
85Grok 4.6High· LMArena data1,463.9
86Qwen3.5 397B A17B· LMArena data1,463.5
87GLM 5· LMArena data1,461.8
88Mistral Medium 3.5· LMArena data1,461.1
89Gemini 3 Flash· LMArena data1,459.6
90Gemma 4 31B IT· LMArena data1,459.3
91Mistral Large 4· LMArena data1,459.2
92Grok 4.20Reasoning, 0309, beta· LMArena data1,458.8
93Grok 4.20 Multi-Agent0309, beta· LMArena data1,457.5
94DeepSeek V4 Flash· LMArena data1,456.5
95Qwen3 Max Preview· LMArena data1,455.9
96Inkling 2 Small· LMArena data1,455.7
97ERNIE 5.00110· LMArena data1,454.7
98GLM 4.7· LMArena data1,454.2
99DeepSeek V3.2Thinking· LMArena data1,453.5
100Gemini 3.5 Flash Lite· LMArena data1,453.5

Showing 51–100 of 418 results.

Data credits: LMArena, CC BY 4.0 (licence)