Loading a pickle that's stored as a row entry?

I am having an issue loading some pickled objects that are stored in an SQL table.

I have a table, tbl, that looks like the following,

In the column 'ModelObject', the strings represent some pickles that were stored as an entry using the function pickle.dumps. I am trying to load the first model object from this SQL table and I cannot seem to get it to load properly. Below is some code that I have tried using to load the pickles but won't work.

conn = pyodbc.connect('Driver={SQL Server};'
                           'Server=MyServer;'
                           'Database=MyDatabase;'
                           'Trusted_Connection=yes;')

sql_query = "SELECT * FROM [MyDatabase].[dbo].[models_data]"

tbl = pd.read_sql(sql_query, conn)

model_1 = pd.read_pickle(tbl.iloc[0]['ModelObject'])
model_1 = pickle.load(tbl.iloc[0]['ModelObject'])

The two last lines both raise different errors, neither working to load the pickles. pd.read_pickle gives the error 'AttributeError: 'bytes' object has no attribute 'seek'' and pickle.load gives the error ''TypeError: file must have 'read' and 'readline' attributes''.

Can anyone help with this?