ModelCards
← All benchmarks

Average

We haven't written a description for this benchmark yet; see the card notes below for how each maker ran it.

Your workloadRAG over docs · 12K in · 600 out · 50K requests/mo · 30% cached
  1. 1Nomic Embed Multimodal 7B · Nomic
    59.7 score
  2. 2jina-embeddings-v5-omni-small · Jina AI
    54.04 score

“Best value”: no other model here reports a higher score at a lower cost for your workload. Scores are self-reported by each maker with different settings, so treat close calls as ties.

How each maker reported it

  • Nomic Embed Multimodal 7B model card performance table

    Nomic's own table, ViDoRe v2 nDCG@5. The card's headline says 58.8 for this model, but its table gives 58.8 to the 3B model and 59.7 to the 7B; the table values are used here.

  • jina-embeddings-v5-omni-small tech report Tables 1–3

    Jina's own numbers on subsets: Text = MMTEB, Image = MIEB, Video = MMEB-Video subset, Audio = MAEB. Parameter counts in the report's tables differ between tables.

Only each model's own maker's number is shown; scores a maker reports for rival models are left out.