Skip to content

Benchmark

Vals Agent Poker Bench

Agentic benchmarks measure how well models carry out tasks that take several steps.

Models measured
17
Scores recorded
17
Sources
1
IQ weight
1.34

Score distribution

1,040.8
Mean
1,038.6
Median
57.9
Std dev
994.5
10th percentile
1,100.2
90th percentile
890.5
Min
890.51,131.8
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.34 for chat models, measured over 17 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard