Skip to content

Benchmark

MRCR 128K (8-needle)

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

Models measured
1
Scores recorded
1
Sources
1
IQ weight
No weight published

Score distribution

90.4%
Mean
90.4%
Median
0%
Std dev
90.4%
10th percentile
90.4%
90th percentile
90.4%
Min

Leaderboard

#ModelScore
1Qwen3.7 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.4%

Data credits: Data by LLM Stats