Benchmark
Arena ELO Vision - Homework
People compare model answers and choose which they prefer. Higher scores are better.
All human preference benchmarksScores from LMArena data.
- Models measured
- 104
- Scores recorded
- 119
- Sources
- 1
- IQ weight
- 2.12
Score distribution
- 1,265.7
- Mean
- 1,276
- Median
- 57.7
- Std dev
- 1,171.2
- 10th percentile
- 1,332.9
- 90th percentile
- 1,117.1
- Min
This benchmark has a weight of 2.12 for chat models, measured over 104 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 | Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data | 1,192.5 |
| 102 | Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data | 1,191.5 |
| 103 | Step3· LMArena data | 1,183.4 |
| 104 | Gemini 2.0 Flash001· LMArena data | 1,176.3 |
| 105 | Mistral Small 3.22506· LMArena data | 1,173.5 |
| 106 | Grok 4.1 FastReasoning· LMArena data | 1,173.3 |
| 107 | Grok 40709· LMArena data | 1,171.2 |
| 108 | Mistral Medium 3.12508· LMArena data | 1,166.7 |
| 109 | Mistral Medium 32505· LMArena data | 1,165.3 |
| 110 | Llama 4 Maverick· LMArena data | 1,164.3 |
| 111 | GLM 4.5V· LMArena data | 1,163.6 |
| 112 | Gemma 3 27B IT· LMArena data | 1,161.4 |
| 113 | GPT-5 nanoHigh· LMArena data | 1,159.6 |
| 114 | Mistral Small 3.1 24B Instruct2503· LMArena data | 1,159.5 |
| 115 | GLM 4.6v· LMArena data | 1,157.4 |
| 116 | Step-1o Turbo202506· LMArena data | 1,153 |
| 117 | Llama 4 Scout· LMArena data | 1,150.4 |
| 118 | Claude 3.5 Sonnet20241022· LMArena data | 1,135.8 |
| 119 | Claude Haiku 3.520241022· LMArena data | 1,117.1 |
Showing 101–119 of 119 results.
Data credits: LMArena, CC BY 4.0 (licence)