Benchmark
Arena ELO Text Factuality - Polish
People compare model answers and choose which they prefer. Higher scores are better.
All human preference benchmarksScores from LMArena data.
- 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
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
| # | Model | Score |
|---|---|---|
| 51 | Gemini 2.5 Flash· LMArena data | 1,408.4 |
| 52 | DeepSeek V3.2· LMArena data | 1,406.7 |
| 53 | Claude Sonnet 4.520250929· LMArena data | 1,406.1 |
| 54 | MiMo V2 FlashNon-Thinking· LMArena data | 1,402.4 |
| 55 | DeepSeek-V4-Flash PreviewHigh· LMArena data | 1,402.2 |
| 56 | Claude Sonnet 4.5High, 32K, 20250929· LMArena data | 1,401.8 |
| 57 | Mistral Medium 3.12508· LMArena data | 1,401.4 |
| 58 | DeepSeek V3.2Thinking· LMArena data | 1,399 |
| 59 | Grok 4.3· LMArena data | 1,396.2 |
| 60 | Qwen3 235B A22B Instruct 25072507· LMArena data | 1,395.6 |
| 61 | Grok 4.1 FastReasoning· LMArena data | 1,394.4 |
| 62 | Step 3.5 Flash· LMArena data | 1,393.4 |
| 63 | Claude Opus 4.120250805· LMArena data | 1,391.4 |
| 64 | MiniMax M2.7· LMArena data | 1,388.7 |
| 65 | GPT-6 LunaMax· LMArena data | 1,388.4 |
| 66 | Kimi K2 Thinking· LMArena data | 1,378.8 |
| 67 | Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data | 1,373.7 |
| 68 | Qwen3.5 Flash· LMArena data | 1,372.5 |
| 69 | Claude Haiku 4.520251001· LMArena data | 1,357.8 |
| 70 | MiniMax M2.5· LMArena data | 1,352.1 |
Showing 51–70 of 70 results.
Data credits: LMArena, CC BY 4.0 (licence)