Benchmark
BLINK - Forensic_Detection
Vision benchmarks measure how well models understand images. Scores are percentages.
All vision benchmarksScores from EvalAI data.
- Models measured
- 4
- Scores recorded
- 4
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 49.2%
- Mean
- 47.7%
- Median
- 7.7%
- Std dev
- 41.7%
- 10th percentile
- 62.1%
- 90th percentile
- 41.7%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Claude 3 OpusForensic_Detection· EvalAI data | 62.1% |
| 2 | Qwen-VL MaxForensic_Detection· EvalAI data | 47.7% |
| 3 | Gemini ProForensic_Detection· EvalAI data | 45.5% |
| 4 | LLaVA 1.6 34BForensic_Detection· EvalAI data | 41.7% |