Skip to content

Benchmark

Protocols Troubleshooting

Science benchmarks measure how well models answer science questions. Scores are percentages.

Models measured
2
Scores recorded
2
Sources
1
IQ weight
No weight published

Score distribution

62.7%
Mean
67.3%
Median
4.7%
Std dev
58%
10th percentile
67.3%
90th percentile
58%
Min
58%67.3%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Claude Sonnet 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim67.3%
2Claude Haiku 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim58%

Data credits: Data by LLM Stats