Skip to content

Benchmark

Arena ELO Text Factuality - Polish

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

Models measured
57
Scores recorded
70
Sources
1
IQ weight
1.96

Score distribution

1,431.3
Mean
1,432.8
Median
31.9
Std dev
1,391.4
10th percentile
1,473.6
90th percentile
1,352.1
Min
1,352.11,488.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.96 for chat models, measured over 57 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 3 Pro Preview· LMArena data1,488.1
2Claude Opus 5Max· LMArena data1,487
3Claude Opus 4.6High· LMArena data1,482.3
4GPT 5.4High· LMArena data1,477
5Gemini 3 Flash· LMArena data1,475.9
6Gemini 3.1 Pro Preview· LMArena data1,475.2
7GPT 5.5High· LMArena data1,475.2
8Claude Opus 5High· LMArena data1,473.6
9Gemini 3.5 FlashMedium· LMArena data1,473
10GPT-5.6 SolxHigh· LMArena data1,470.5
11Claude Opus 4.7· LMArena data1,470.2
12GPT 5.5· LMArena data1,465.6
13Claude Opus 4.6· LMArena data1,465.5
14Claude Opus 4.7High· LMArena data1,463.5
15Claude Opus 4.8High· LMArena data1,457
16Claude Opus 4.8· LMArena data1,456.2
17Gemini 3.8 FlashHigh· LMArena data1,456
18Gemini 3 FlashMinimal· LMArena data1,455.8
19Gemini 2.5 Pro· LMArena data1,455.3
20GPT 5.4· LMArena data1,454.7
21MiMo V2.5 Pro· LMArena data1,451.8
22Grok 4.5· LMArena data1,451.2
23GPT-5.6 TerraxHigh· LMArena data1,445.9
24GPT 5.1· LMArena data1,445.1
25ChatGPT-4o Latest20250326· LMArena data1,444.3
26Qwen3.5 397B A17B· LMArena data1,441.7
27GPT 5.2· LMArena data1,439.9
28GLM 5.1· LMArena data1,439.5
29Grok 4.20 Multi-Agent0309, beta· LMArena data1,438.8
30Doubao Seed 2.0 Pro· LMArena data1,438.7
31GPT 5.1High· LMArena data1,438
32Qwen3.7 Plus· LMArena data1,437.5
33Claude Opus 4.520251101· LMArena data1,436.2
34Gemini 3.1 Flash Lite· LMArena data1,432.9
35GPT-5.4 miniHigh· LMArena data1,432.8
36GLM 5.3 Flash· LMArena data1,432.6
37DeepSeek-V4-Pro PreviewHigh· LMArena data1,431.6
38Inkling FP4· LMArena data1,431.3
39Grok 4.20Reasoning, 0309, beta· LMArena data1,431
40Claude Opus 4.5High, 32K, 20251101· LMArena data1,430.6
41GLM 5.2Max· LMArena data1,429.8
42ERNIE 5.00110· LMArena data1,428.8
43Claude Sonnet 4.6· LMArena data1,426.1
44GPT 5.2High· LMArena data1,426
45Grok 4.1Thinking· LMArena data1,425.9
46Kimi K2.5Thinking· LMArena data1,423.3
47MiniMax M3· LMArena data1,422.7
48Grok 4.1· LMArena data1,420.1
49Mistral Large 3· LMArena data1,417.6
50GLM 4.6· LMArena data1,414.2

Showing 1–50 of 70 results.

Data credits: LMArena, CC BY 4.0 (licence)