Review

http://www.mirrorlessons.com/2016/07/15/panasonic-leica-12mm-f1-4-review/

Interesting that that at about F4 the Olympus is sharper, although I would not be inclined to go by mtf numbers alone. Very impressive numbers for F2 from the Panasonic, though.

http://pen-and-tell.blogspot.com/2016/07/neuer-video-und-das-panasonic-12mm-14.html

@apefos et al.
It's quite common to see wildly varying measurements in reviews (look at the Sigma 60mm, it's insane). It's hard to say exactly why this is, but here's what we know:

1. Sample variation. It can be quite noticeable, the difference in copies of the same lens. LensRentals usually tests four or five copies, the other sites really do not, so you can basically factor in 10-15 percent in the results right there--you shouldn't buy a lens based on a small difference from a reviewer that does not test several copies.
2. Pilot error. In quite a few of these tests, the focus point is slightly off. Ouch. Typically, they take a dozen or so shots and then pick the "best" ones, but often there are problems. A millimeter makes a big difference at 50 feet, if you are looking at landscapes.
3. The camera. Different testers use different cameras. For example, the very well done test at cameralabs inexplicably uses the 16mm sensor camera. Now, there is no AA filter, but this is not the absolute best camera for stills, and, also, you are just going to get different results from different cameras. The GX85 does not suffer as much from shutter shock, at least. When you a re looking at tiny differences, the camera can make a difference, Does anyone think to try several cameras? No. Obviously, you can optimize the camera settings for each lens, but no one ever does this because they think, incorrectly, that all settings should be the same. Ideally, the lens should be tested both ways--once to see what the differences are, and then a second time with optimized settings to see what you can squeeze out of it, which is what you will be doing in real life.
4. The subject material...different subjects reveal different strengths in the lens, you could move the whole setup to a different venue and get different results. Also, each lens has an ideal distance, so in one test they are shooting at ten feet, and in another they are shooting at 50 feet--this will give you differences, they should measure at several distances, or at least try to determine the optimum distance for each lens.
5. Burned cookies. We never see, of the dozens of shots for the "test" the bad ones. They have just as much info as the good ones, maybe more.

In stock at

https://www.amazon.com/Panasonic-SUMMILUX-Professional-Mirrorless-H-X012/dp/B01H1MNW68/

I don't like the bokeh--it's edgy, especially on parts of the background that have some detail. Look at the upper left corner of the photo, it's edgy, it's not still, it's turbulent. Even an old minolta lens has better bokeh. One of the reasons is that the lens is hyperfocal, of course.
If you are buying this lens, you aren't buying it for the bokeh anyway, there are many better choices for DOF.

Summilux delivered sharp results into the corners even with the aperture wide-open. The optical and build quality are a step-up from the Olympus 12mm f2, but equally it's comfortably larger, heavier and a lot more expensive too.

http://www.cameralabs.com/reviews/Panasonic_Leica_DG_Summilux_12mm_f1-4_H-X012/

Review

https://blog.mingthein.com/2018/02/25/review-the-panasonic-leica-12mm-f1-4-asph-for-micro-four-thirds/