Benchmark
MMEB Image - DocVQA
Vision benchmarks measure how well models understand images. Scores are percentages.
- Models measured
- 1
- Scores recorded
- 1
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 96.3%
- Mean
- 96.3%
- Median
- 0%
- Std dev
- 96.3%
- 10th percentile
- 96.3%
- 90th percentile
- 96.3%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Qwen3-VL Embedding 8BDocVQA· MMEB (TIGER-Lab) data | 96.3% |
Data credits: MMEB Leaderboard (TIGER-Lab), Apache-2.0 (licence)