Skip to content

Benchmark

Agentset Embedding nDCG@10 - ARCD

Embedding benchmarks measure how well models find related items, for example when matching a search to relevant results. Scores are percentages.

Models measured
15
Scores recorded
15
Sources
1
IQ weight
1.00

Score distribution

79.9%
Mean
86.4%
Median
21%
Std dev
76.3%
10th percentile
90.5%
90th percentile
2.7%
Min
2.7%91.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.00 for embedding models, measured over 15 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard