Skip to content

Benchmark

Arena ELO Text Factuality - Instruction Following

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

Models measured
160
Scores recorded
179
Sources
1
IQ weight
2.15

Score distribution

1,436.3
Mean
1,439.3
Median
44.1
Std dev
1,377.3
10th percentile
1,486.3
90th percentile
1,280.8
Min
1,280.81,523.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.15 for chat models, measured over 160 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 data1,523.5
2Claude Opus 5.5High· LMArena data1,522.4
3Gemini 4 NeonHigh· LMArena data1,521.5
4Gemini 4 HydrogenHigh· LMArena data1,512.1
5Claude Fable 5.1Max· LMArena data1,511.4
6Gemini 4 ArgonHigh· LMArena data1,510
7Claude Opus 4.6High· LMArena data1,508.5
8Claude Opus 5Max· LMArena data1,501.1
9Claude Opus 4.6· LMArena data1,499.1
10Claude Fable 5.2Max· LMArena data1,495.9
11Claude Opus 5High· LMArena data1,494.6
12Claude Fable 5High· LMArena data1,494.4
13Kimi K3.1 ProMax· LMArena data1,493.5
14Muse WonderMax· LMArena data1,492.3
15GPT-5.6 SolxHigh· LMArena data1,491.1
16GPT 5.5High· LMArena data1,490.6
17GPT 5.5· LMArena data1,488.6
18Claude Opus 4.7High· LMArena data1,488
19MiMo V2.6 Pro· LMArena data1,486.3
20Claude Sonnet 5.5xHigh· LMArena data1,484.3
21Claude Opus 4.5High, 32K, 20251101· LMArena data1,483.8
22GPT 5.4High· LMArena data1,483.7
23Claude Opus 4.7· LMArena data1,483
24Claude Opus 4.8High· LMArena data1,482.9
25GPT-6.1 SolMax· LMArena data1,481.2
26Muse Spark 1.3Max· LMArena data1,481
27Kimi K3Max· LMArena data1,480.5
28Qwen3.5-Max-Preview· LMArena data1,479
29Qwen3.8 Max· LMArena data1,478.8
30Gemini 3.7 FlashHigh· LMArena data1,478.1
31DeepSeek V4.1 Flash 0910Max· LMArena data1,478
32GLM 5.3Max· LMArena data1,477.7
33Claude Opus 4.520251101· LMArena data1,477.4
34Claude Sonnet 4.6· LMArena data1,476
35Qwen3.7 Max Preview· LMArena data1,474.7
36Gemini 3.8 FlashHigh· LMArena data1,474.7
37MiMo V2.5 Pro· LMArena data1,474.3
38GLM 5.3 Flash· LMArena data1,474.2
39GPT 5.4· LMArena data1,474.1
40GPT-5.6 TerraxHigh· LMArena data1,471.5
41Gemini 3.5 FlashHigh· LMArena data1,471.3
42Muse Spark 1.2xHigh· LMArena data1,471
43GPT-6 AstraMax· LMArena data1,470.4
44Claude Opus 4.8· LMArena data1,470.3
45GLM 5.2Max· LMArena data1,469
46Gemini 3.6 FlashHigh· LMArena data1,469
47Gemini 3 Pro Preview· LMArena data1,468.7
48Claude Sonnet 4.520250929· LMArena data1,468.3
49Gemini 3.1 Pro Preview· LMArena data1,466.4
50Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,466

Showing 1–50 of 179 results.

Data credits: LMArena, CC BY 4.0 (licence)