Skip to content

Benchmark

GDM-MRCRv2: 8 needle 512k

Long context benchmarks measure how well models use information in long inputs. Scores are percentages.

Models measured
40
Scores recorded
107
Sources
1
IQ weight
1.96

Score distribution

31.4%
Mean
29.7%
Median
15.4%
Std dev
13.2%
10th percentile
52.5%
90th percentile
7.3%
Min
7.3%77.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.96 for chat models, measured over 40 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

Showing 101–107 of 107 results.