Benchmark
Arena ELO Vision - Creative Writing Vision
People compare model answers and choose which they prefer. Higher scores are better.
All human preference benchmarksScores from LMArena data.
- Models measured
- 93
- Scores recorded
- 106
- Sources
- 1
- IQ weight
- 1.74
Score distribution
- 1,246.2
- Mean
- 1,254
- Median
- 53.6
- Std dev
- 1,181.7
- 10th percentile
- 1,318.8
- 90th percentile
- 1,100
- Min
This benchmark has a weight of 1.74 for chat models, measured over 93 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 | Hunyuan Vision 1.5 Thinking· LMArena data | 1,146.7 |
| 102 | Mistral Medium 32505· LMArena data | 1,146.6 |
| 103 | Llama 4 Scout· LMArena data | 1,127.2 |
| 104 | Mistral Small 3.1 24B Instruct2503· LMArena data | 1,124 |
| 105 | GLM 4.6v· LMArena data | 1,113.9 |
| 106 | Llama 4 Maverick· LMArena data | 1,100 |
Showing 101–106 of 106 results.
Data credits: LMArena, CC BY 4.0 (licence)