Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ChemBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
General and STEM reasoningChemBench
Pass@167.86
20
Scientific ReasoningChemBench (test)
Initial Score66
13
Chemical knowledge and reasoningChemBench
Overall Score86.1
11
Technical chemistryChemBench Technical chemistry
Spearman Correlation0.86
8
Physical chemistryChemBench Physical chemistry
Spearman Correlation0.74
8
Organic chemistryChemBench Organic chemistry
Spearman Correlation0.89
8
Material scienceChemBench Material science
Spearman Correlation0.42
8
Inorganic chemistryChemBench Inorganic chemistry
Spearman Correlation0.67
8
Analytical chemistryChemBench Analytical chemistry
Spearman Correlation0.86
8
Chemical reasoning and predictionChemBench4K
Product93.66
7
In-Context LearningChemBench
Accuracy58.4
6
ScienceChemBench
Score83.3
3
Chemistry Question AnsweringChemBench Overall
Success90.5
1
Chemistry Question AnsweringChemBench Toxicity and Safety
Success Rate90.1
1
Chemistry Question AnsweringChemBench Technical Chemistry
Success Rate0.825
1
Chemistry Question AnsweringChemBench Physical Chemistry
Success8,420
1
Chemistry Question AnsweringChemBench Organic Chemistry
Success92.1
1
Chemistry Question AnsweringChemBench Materials Science
Success Rate90.5
1
Chemistry Question AnsweringChemBench Inorganic Chemistry
Success Rate91.3
1
Chemistry Question AnsweringChemBench General Chemistry
Success Rate0.872
1
Chemistry Question AnsweringChemBench Chemical Preference
Success Rate93.1
1
Chemistry Question AnsweringChemBench Analytical Chemistry
Success Rate82.2
1
Showing 22 of 22 rows