Benchmark
ViDoRe(v3.1) - Private split
Vision benchmarks measure how well models understand images. Scores are percentages.
- Models measured
- 9
- Scores recorded
- 9
- Sources
- 1
- IQ weight
- 1.37
Score distribution
- 49.6%
- Mean
- 53.8%
- Median
- 9.5%
- Std dev
- 44.1%
- 10th percentile
- 56.7%
- 90th percentile
- 26.5%
- Min
This benchmark has a weight of 1.37 for embedding models, measured over 9 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 | Nemotron 3.5 Embed 8bPrivate split· MTEB Multimodal Boards data docs | 60% |
| 2 | Nemotron 3.5 Embed 1bPrivate split· MTEB Multimodal Boards data docs | 56.7% |
| 3 | Nemotron 3 Embed 1bPrivate split· MTEB Multimodal Boards data docs | 55% |
| 4 | Qwen3-VL Embedding 8BPrivate split· MTEB Multimodal Boards data docs | 54.9% |
| 5 | Jina Embeddings v4Private split· MTEB Multimodal Boards data docs | 53.8% |
| 6 | Llama Nemotron Embed VL 1b V2Private split· MTEB Multimodal Boards data docs | 50.9% |
| 7 | Jina Embeddings v5 Omni NanoPrivate split· MTEB Multimodal Boards data docs | 45% |
| 8 | Jina Embeddings v5 Omni SmallPrivate split· MTEB Multimodal Boards data docs | 44.1% |
| 9 | bge-large-enPrivate split· MTEB Multimodal Boards data docs | 26.5% |