The coefficients can be obtained using the params attribute of a fitted model.
Consider the following code:
import statsmodels.api as sm
import numpy as np
np.random.seed(1)
val_x = sm.add_constant(np.arange(100))
val_ y = np.dot(val_x, [1,2]) + np.random.normal(size=100)
output = sm.OLS(val_y, val_x).fit()
print(output.params)
It will print a numpy array with the values [0.89516052 2.00334187] for the intercept and slope, respectively.
If you need more information, utilize the output.summary() object, which has three full tables with model descriptions.