jina-embeddings-v5-omni: Text-Geometry-Preserving Multimodal Embeddings via Frozen-Tower Composition
Open paper
Benchmark
Model
Rank
Results
visual-document-retrieval-on-vidore-v3
jina-embeddings-v5-omni-small
#4
NDCG@10: 45.71
representation-learning-on-mteb-english-v2
jina-embeddings-v5-omni-small
#11
MTEB Score: 71.7800
Rank counts only results with a code link.