MTBench
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MTbench (test)
2.24StdDev
45
MTBench
9.14MTBench Score
43
MTBench101
9.03Score
33
MTBench
9.35Helpfulness
18
MM MTBench
84.9Overall Score
18
MTBench Temperature Forecasting (14-day)
5.026MSE
17
MTBench Stock Price Forecasting (30-day)
1.122MAE
17
MTBench Stock Price Forecasting 7-day
0.788MAE
17
MTBench Human
88.9Accuracy
16
MTBench MathTutorBench OOD
8.29Score (Sc)
13
MTBench Weather (Long)
93.496Past Accuracy
10
MTBench Weather Short
93.877Past Trend Prediction Score
10
MTBench Finance Long
62.6713-way Accuracy
10
MTBench Finance Short
66.8493-way Score
10
MTBench Weather Long
11.823MSE
10
MTBench Weather Short
10.02MSE
10
MTBench Finance (Long)
3.531MAPE
10
MTBench Finance Short
2.545MAPE
10
MTBench Weather Indicator
10.898Max MSE
9
MTBench 7-day weather forecast
10.719MSE
9
MTBench Weather
71.7Accuracy
9
MTBench Finance
91.3Accuracy
9
MTBench Weather
3.523MAE
9
MTBench Finance
0.814MAE
9
MTBench Weather
55.7Accuracy
9