https://en.wikipedia.org/wiki/Opinion_polling_for_the_United_Kingdom_general_election,_2017
You can sort by poll name too, to see the trend in each. Different pollsters have used different ways to estimate the likelihood of voting, especially youth, and that's been responsible for a lot of the differences between them.
For seat estimates, I know of 2:
YouGov, a pollster
https://yougov.co.uk/uk-general-election-2017/ - estimating Con 302, Lab 269, SNP 44, Lib Dem 12, Plaid Cymru 2, Green 1, 'Other' 2 (and Northern Ireland 18 - completely different parties, so done separately, though the SDLP would often vote with Labour). That 'Other' consists of the Speaker of the Commons - traditionally unopposed by major parties, and a weird result for East Devon of a win for an independent county councillor their model comes up with (which, last I looked, wasn't expected in the local press at all - which may show the YouGov model has significant flaws). Model explained here:
https://yougov.co.uk/news/2017/06/07/two-methods-one-commitment-yougovs-polling-and-mod/
And
http://www.electoralcalculus.co.uk/homepage.html - Con 361, Lab 216, SNP 48, Lib Dem 3 ... . That model is explained here:
https://www.theguardian.com/science/2017/apr/27/lib-dems-shouldnt-count-on-remain-votes-the-data-looks-bleak . That pair think the Tories are doing better in the marginals, hence the far larger lead they give them. But you can see there's a lot of disagreement.
= new reply since forum marked as read
