Benchmark
Arena ELO Text Factuality - Multi Turn
People compare model answers and choose which they prefer. Higher scores are better.
All human preference benchmarksScores from LMArena data.
- Models measured
- 142
- Scores recorded
- 161
- Sources
- 1
- IQ weight
- 2.11
Score distribution
- 1,450.4
- Mean
- 1,457.6
- Median
- 36.7
- Std dev
- 1,398.3
- 10th percentile
- 1,490.7
- 90th percentile
- 1,341.4
- Min
This benchmark has a weight of 2.11 for chat models, measured over 142 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 |
|---|---|---|
| 151 | GLM 4.5 Air· LMArena data | 1,381.4 |
| 152 | Trinity Large Preview· LMArena data | 1,378.5 |
| 153 | GPT OSS 120B· LMArena data | 1,377.1 |
| 154 | nova-experimental-chat10-20· LMArena data | 1,376.8 |
| 155 | Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data | 1,367.6 |
| 156 | GLM 4.7 Flash· LMArena data | 1,367.5 |
| 157 | Nova 2 Lite· LMArena data | 1,359 |
| 158 | Trinity Large Thinking· LMArena data | 1,355.8 |
| 159 | Command A03-2025· LMArena data | 1,347.6 |
| 160 | Olmo 3.1 32B Instruct· LMArena data | 1,346.1 |
| 161 | Nemotron 3 Nano 30B A3BBF16· LMArena data | 1,341.4 |
Showing 151–161 of 161 results.
Data credits: LMArena, CC BY 4.0 (licence)