Benchmark
Arena ELO Text Factuality - Industry Mathematical
People compare model answers and choose which they prefer. Higher scores are better.
All human preference benchmarksScores from LMArena data.
- Models measured
- 102
- Scores recorded
- 117
- Sources
- 1
- IQ weight
- 2.25
Score distribution
- 1,453.4
- Mean
- 1,452.4
- Median
- 30.2
- Std dev
- 1,416.5
- 10th percentile
- 1,489.5
- 90th percentile
- 1,363.5
- Min
This benchmark has a weight of 2.25 for chat models, measured over 102 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 |
|---|---|---|
| 101 | Gemini 2.5 Flash· LMArena data | 1,420.7 |
| 102 | nova-experimental-chat11-10· LMArena data | 1,420.4 |
| 103 | GPT-5.5 Instant· LMArena data | 1,417.8 |
| 104 | Qwen3.5 35B A3B· LMArena data | 1,417.1 |
| 105 | Grok 4.1Thinking· LMArena data | 1,416.5 |
| 106 | ChatGPT-4o Latest20250326· LMArena data | 1,416.3 |
| 107 | Qwen3.5 Flash· LMArena data | 1,412.7 |
| 108 | Doubao Seed 2.0 Pro· LMArena data | 1,411.8 |
| 109 | Grok 4.3· LMArena data | 1,411.2 |
| 110 | Grok 4.1 FastReasoning· LMArena data | 1,402.9 |
| 111 | GPT-5.3 Chat· LMArena data | 1,402.4 |
| 112 | Grok 4.1· LMArena data | 1,401.4 |
| 113 | MiMo V2 FlashNon-Thinking· LMArena data | 1,400.8 |
| 114 | Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data | 1,399.2 |
| 115 | MiniMax M2.5· LMArena data | 1,389 |
| 116 | Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data | 1,387.4 |
| 117 | Trinity Large Preview· LMArena data | 1,363.5 |
Showing 101–117 of 117 results.
Data credits: LMArena, CC BY 4.0 (licence)