Skip to content

Benchmark

Arena ELO Text Factuality - Non English

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
168
Scores recorded
192
Sources
1
IQ weight
2.36

Score distribution

1,422.4
Mean
1,430.7
Median
42.6
Std dev
1,361
10th percentile
1,469
90th percentile
1,245.6
Min
1,245.61,512.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.36 for chat models, measured over 168 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51GLM 5.2Max· LMArena data1,449.4
52MiMo V2.5 Pro· LMArena data1,446.9
53GPT-5.6 LunaxHigh· LMArena data1,444.8
54Claude Opus 4.520251101· LMArena data1,444.4
55GPT 5.1High· LMArena data1,444.4
56Gemini 3 FlashMinimal· LMArena data1,444.1
57Grok 4.20 Multi-Agent0309, beta· LMArena data1,443.8
58Kimi K2.6· LMArena data1,443.8
59Claude Opus 4.5High, 32K, 20251101· LMArena data1,443.6
60GPT-6.1 TerraMax· LMArena data1,443.6
61Grok 4.20Reasoning, 0309, beta· LMArena data1,442.9
62Claude Sonnet 4.6· LMArena data1,442.5
63ERNIE 5.1· LMArena data1,442.3
64Inkling FP4· LMArena data1,441.8
65DeepSeek V4 Pro· LMArena data1,441.5
66GLM 5.1· LMArena data1,440.9
67Qwen3.6 Max Preview· LMArena data1,439.9
68Claude Sonnet 5High· LMArena data1,439.4
69Grok 4.6High· LMArena data1,439.2
70GPT-6.2 SolMax· LMArena data1,438.5
71Grok 4.20 Beta 1· LMArena data1,438.4
72Qwen3.7 Plus· LMArena data1,438.3
73DeepSeek V4 ProHigh, 20260813· LMArena data1,436.6
74Gemini 3.5 Flash Lite· LMArena data1,436.5
75GPT-5.4 miniHigh· LMArena data1,436.1
76GLM 5· LMArena data1,435.9
77Gemma 4 31B IT· LMArena data1,435.2
78GPT-5.2 Chat20260210· LMArena data1,435
79GPT-5.5 Instant· LMArena data1,434.9
80nova-experimental-chat26-02-10· LMArena data1,434.7
81Qwen3.5 397B A17B· LMArena data1,434.7
82Hy3 (Deep Infra)· LMArena data1,434.1
83Claude Sonnet 4.520250929· LMArena data1,434
84Step 5 Preview· LMArena data1,433.9
85ERNIE 5.00110· LMArena data1,433.7
86Qwen3 Max Preview· LMArena data1,433.2
87ChatGPT-4o Latest20250326· LMArena data1,432.8
88ERNIE-5.0-Preview1203· LMArena data1,432.5
89GPT 5.1· LMArena data1,432.5
90GLM 4.6· LMArena data1,432
91DeepSeek-V4-Pro PreviewHigh· LMArena data1,431.8
92ERNIE-5.0-Preview1022· LMArena data1,431.5
93MiMo V2.6 Flash· LMArena data1,431.2
94GPT-6.2 AstraMax· LMArena data1,431.1
95Gemma 4 26B A4B· LMArena data1,431
96Kimi K2.5Thinking· LMArena data1,430.9
97GPT 5.2· LMArena data1,430.7
98GPT-6 SolMax· LMArena data1,429.4
99Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,429.1
100GPT 5.2High· LMArena data1,427.6

Showing 51–100 of 192 results.

Data credits: LMArena, CC BY 4.0 (licence)