aiaimodel.cards
← All benchmarks

VMQA/VCT

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. 1Claude Haiku 4.5 · Anthropicbest value
    0.32 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

  • Claude Haiku 4.5 §6, p.37–39

    The card has no general capability table; it points to the launch post for benchmarks. These are the Responsible Scaling Policy dangerous-capability evaluations it reports, from the highest-scoring (most capable) snapshot. Scores are 0–1 unless marked.

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