Power users may want to export raw data from modelsummary for further post-processing before creating their tables. They can achieve that by setting the output argument to data.frame:

library(modelsummary)
mod <- lm(mpg ~ hp, mtcars)
modelsummary(mod, output="data.frame")
## # A tibble: 11 x 4
##    group     term          statistic    `Model 1`
##    <chr>     <chr>         <chr>        <chr>    
##  1 estimates "(Intercept)" "estimate"   30.099   
##  2 estimates ""            "statistic1" (1.634)  
##  3 estimates "hp"          "estimate"   -0.068   
##  4 estimates ""            "statistic1" (0.010)  
##  5 gof       "Num.Obs."    ""           32       
##  6 gof       "R2"          ""           0.602    
##  7 gof       "R2 Adj."     ""           0.589    
##  8 gof       "AIC"         ""           181.2    
##  9 gof       "BIC"         ""           185.6    
## 10 gof       "Log.Lik."    ""           -87.619  
## 11 gof       "F"           ""           45.460