Late reply here but no. I did not consider it. It could be possible I think.
we have only English version and our offline copies has really big difference in weight, why? i mean 32 mb of mine and 76 mb of yours
I believe mine is scraping more images. I simply scrape every page branching off a URL. I branch from the main page, http://wiki.multitheftauto.com/wiki/Main_Page.
That does not include history as I mentioned. You have to start from a different URL to scrape that and it is massive uncompressed. Like I stopped at 800MB IIRC. That link is a higher branch, http://wiki.multitheftauto.com/index.php.
I don't scrape other languages, that is also a different URL branch. PL example: http://wiki.multitheftauto.com/index.php?title=PL/Main_Page.
I am using Teleport VLX. Before I was using another version that would max out at 65,535 URLs although that covered all the critical stuff and the actual tally is not too far above that.