Skip to content

Benchmark

Arena ELO Text Factuality - Coding

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

Models measured
155
Scores recorded
174
Sources
1
IQ weight
1.95

Score distribution

1,491.4
Mean
1,494.7
Median
38
Std dev
1,437
10th percentile
1,534.3
90th percentile
1,344.9
Min
1,344.91,567
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.95 for chat models, measured over 155 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 4 NeonHigh· LMArena data1,567
2Gemini 4 CarbonHigh· LMArena data1,565.4
3Claude Opus 5.5High· LMArena data1,561.5
4Claude Opus 4.6High· LMArena data1,550.9
5MiMo V2.6 Pro· LMArena data1,548
6Claude Opus 4.6· LMArena data1,547.3
7Claude Opus 5Max· LMArena data1,546.3
8Gemini 4 ArgonHigh· LMArena data1,546
9Muse WonderMax· LMArena data1,544.9
10Claude Fable 5.1Max· LMArena data1,543.3
11Claude Fable 5High· LMArena data1,536.8
12Kimi K3.1 ProMax· LMArena data1,536.1
13Gemini 4 HydrogenHigh· LMArena data1,535.6
14Claude Opus 5High· LMArena data1,535.6
15Muse Spark 1.3Max· LMArena data1,535.5
16GPT-6 AstraMax· LMArena data1,535.1
17Claude Opus 4.7High· LMArena data1,534.6
18GPT 5.4High· LMArena data1,534.3
19GPT-6.1 SolMax· LMArena data1,534.2
20Claude Opus 4.5High, 32K, 20251101· LMArena data1,533.5
21Claude Opus 4.7· LMArena data1,532.5
22GPT-5.6 SolxHigh· LMArena data1,532.4
23GPT 5.5High· LMArena data1,532.2
24Claude Opus 4.520251101· LMArena data1,531.2
25Kimi K3Max· LMArena data1,530.3
26Claude Sonnet 4.6· LMArena data1,529.3
27DeepSeek V4.1 Flash 0910Max· LMArena data1,529.3
28Muse Spark 1.2xHigh· LMArena data1,529.1
29GPT-6.1 TerraMax· LMArena data1,528.8
30MiMo V2.5 Pro· LMArena data1,527.9
31GLM 5.3 Flash· LMArena data1,526.8
32Claude Sonnet 5.5xHigh· LMArena data1,526.4
33GPT 5.4· LMArena data1,524.9
34GPT 5.5· LMArena data1,524.8
35Qwen3.5-Max-Preview· LMArena data1,524.7
36Qwen3.8 Max· LMArena data1,524.5
37Gemini 3.7 FlashHigh· LMArena data1,524.3
38GLM 5.3Max· LMArena data1,524.1
39Claude Opus 4.8High· LMArena data1,523.6
40GPT-5.6 TerraxHigh· LMArena data1,522.4
41Claude Sonnet 4.520250929· LMArena data1,522.1
42Gemini 3.5 FlashHigh· LMArena data1,520.9
43Beam 1.0 501B A23B· LMArena data1,520.1
44Gemini 3.6 FlashHigh· LMArena data1,519.8
45MiMo V2.6 Flash· LMArena data1,519.4
46Claude Opus 4.8· LMArena data1,519.4
47Gemini 3 Pro Preview· LMArena data1,519.2
48Gemini 3.8 FlashHigh· LMArena data1,518.7
49Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,518.4
50GPT-6 LunaMax· LMArena data1,518.4

Showing 1–50 of 174 results.

Data credits: LMArena, CC BY 4.0 (licence)