Skip to content

Benchmark

Arena ELO - Instruction Following

Instruction following benchmarks measure how closely models follow instructions. Higher scores are better.

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.18

Score distribution

1,294.5
Mean
1,332.7
Median
155.5
Std dev
1,056.7
10th percentile
1,465
90th percentile
780.7
Min
780.71,539.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.18 for chat models, measured over 366 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 CarbonHigh· LMArena data1,539.1
2Gemini 4 NeonHigh· LMArena data1,535.3
3Gemini 4 ArgonHigh· LMArena data1,535.1
4Claude Opus 5.5High· LMArena data1,532.6
5Claude Fable 5.1Max· LMArena data1,521.6
6Claude Opus 4.6High· LMArena data1,521.2
7Claude Opus 5Max· LMArena data1,517.5
8Gemini 4 HydrogenHigh· LMArena data1,515
9Claude Opus 5High· LMArena data1,514.9
10Claude Opus 4.6· LMArena data1,510
11Claude Fable 5High· LMArena data1,502.6
12Claude Opus 4.7High· LMArena data1,498.4
13Gemini 3.8 FlashHigh· LMArena data1,494.2
14MiMo V2.6 Pro· LMArena data1,493.8
15Muse WonderMax· LMArena data1,491.7
16Claude Fable 5.2Max· LMArena data1,490.1
17Claude Opus 4.7· LMArena data1,488.9
18Gemini 3.7 FlashHigh· LMArena data1,485.2
19Claude Sonnet 5.5xHigh· LMArena data1,484.7
20Kimi K3Max· LMArena data1,483.9
21Qwen3.8 Max· LMArena data1,483.7
22GPT-5.6 SolxHigh· LMArena data1,482.2
23GLM 5.3Max· LMArena data1,482.1
24GLM 5.3 Flash· LMArena data1,480.2
25Muse Spark 1.3Max· LMArena data1,480
26GPT 5.5High· LMArena data1,477.9
27Claude Opus 4.5High, 32K, 20251101· LMArena data1,477.8
28Claude Opus 4.8High· LMArena data1,477.2
29MiMo V2.5 Pro· LMArena data1,475.6
30Claude Sonnet 4.6· LMArena data1,475.5
31GPT 5.5· LMArena data1,474.7
32Kimi K3.1 ProMax· LMArena data1,473.4
33Claude Opus 4.520251101· LMArena data1,472.5
34DeepSeek V4.1 Flash 0910Max· LMArena data1,472.4
35GPT 5.4High· LMArena data1,469.2
36Qwen3.5-Max-Preview· LMArena data1,468
37Gemini 3.5 FlashHigh· LMArena data1,467.6
38Gemini 3.6 FlashHigh· LMArena data1,467.3
39Muse Spark 1.2xHigh· LMArena data1,466.8
40GLM 5.2Max· LMArena data1,466.7
41GPT-6.1 SolMax· LMArena data1,465.9
42Gemini 3.1 Pro Preview· LMArena data1,465.3
43Gemini 3.5 FlashMedium· LMArena data1,465
44Claude Opus 4.8· LMArena data1,462.9
45ERNIE 5.1· LMArena data1,461.7
46Qwen3.7 Max Preview· LMArena data1,460
47Claude Sonnet 4.520250929· LMArena data1,459
48Gemini 3 Pro Preview· LMArena data1,458
49Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,457.8
50Muse Spark 1.1· LMArena data1,457.4

Showing 1–50 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)