Is there a way to load a file with a different number of columns in each row into the read_table in Pandas?

For example,

1 2
2 3 4
3

When I tried to load a file like this into read_table of pandas, I saw 12 in the first row and automatically recognized that there were two line columns, so the second row with 234 showed a tokenizing error.

There are 24 text files that need to be retrieved, but it takes a long time to open one text file beyond 1GB, so it is difficult to bring up rows with the maximum column one by one.

I'm worried about using the bad_line_error = False option because I need all the data.

When I tried to list it as open().read().splitlines and put it in the dataframe, the code was interrupted due to insufficient memory. (Of course, I got the 64 bit version)

I wish I could set the number of columns in advance, but I can't think of a way. If anyone knows, please give me some advice.