Benchmark
Vals CUA-bench
Vision benchmarks measure how well models understand images. Scores are percentages.
- Models measured
- 8
- Scores recorded
- 8
- Sources
- 1
- IQ weight
- 0.45
Score distribution
- 10%
- Mean
- 9%
- Median
- 4.8%
- Std dev
- 4.8%
- 10th percentile
- 14%
- 90th percentile
- 4.2%
- Min
This benchmark has a weight of 0.45 for chat models, measured over 8 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 |
|---|---|---|
| 1 | GPT-6 Astra· Vals AI | 19.2% |
| 2 | Claude Opus 5.5· Vals AI | 14% |
| 3 | Claude Fable 5.1· Vals AI | 13.2% |
| 4 | Claude Opus 5· Vals AI | 9% |
| 5 | GPT-5.6 Sol· Vals AI | 8.3% |
| 6 | Muse Wonder· Vals AI | 7.2% |
| 7 | Gemini 4 Argon· Vals AI | 4.8% |
| 8 | Gemini 3.8 Flash· Vals AI | 4.2% |