Skip to content

Benchmark

MedChemBench (Internal)

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

Models measured
4
Scores recorded
4
Sources
1
IQ weight
No weight published

Score distribution

35.3%
Mean
35%
Median
8%
Std dev
27.5%
10th percentile
48.3%
90th percentile
27.5%
Min
27.5%48.3%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1GPT-5.6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim48.3%
2GPT-5.6 Terra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim35%
3GPT-5.6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim30.4%
4GPT-rosalind-research· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim27.5%

Data credits: Data by LLM Stats