Without extensive dynamometer testing to compare actual numbers and quantify all the different tunes available both DIY and GDE, any comparison would be purely speculative at this point... All of them have them have their users who swear by this one or that one being the best available, but at the end of the day price may dictate a drivers decision on what they are willing or not willing to spend to gain similar results.....

Sure, scientific lab based results would be more accurate but there are claims of 10% or more improvement which should be noticeable by someone driving to work the same as they always do, etc. There are important factors other than price such as convenience. Having to send an ECU in to get modified or spend $200 more for a device that may be used only once doesn't seem very logical.
So, what do you or have you run for ECU and what is your real life experience?

Any of the tunes can increase the HP & torque which will require an update of your torque converter to at least a HEMI or Suncoast TC to be able to handle the additional HP & torque. As to mileage; depends on the driver and can he/she keep their foot out of the throttle during acceleration since it runs so good?

I can only speak for myself; I am running the DIY Stage IV Hot Tune and on the interstate holding constant speeds I can squeeze ~37 mpg out of it, but you have to drive it like you got an egg under your foot. And I have repeated this MPG on more than one occasion so I know it is repeatable. Around town, high 20's are the best I have been able to achieve on a consistent basis...

But to be honest, most times I say the heck with MPG and drive it like the little Italian race car that it is! It really likes to be driven with gusto and it feels good to have all that power under your foot!!!!!

Yeah, its hard to compare the two from a performance/mpg perspective, but I would guess that the GDE tunes are probably far more comprehensive than what most of the DIY tunes offer. They know what and where all the maps are in the tune and have the equipment to test/verify their changes have the intended effect. You can be certain that they are well tested and do what they say. Thats why they cost money.

Although we have a pretty good understanding of how the tunes are laid out and what each map does, there are still certain things that are outside of our understanding. Its easy to turn up the fueling, but things like turbo VNT control, injection timing, # of injection events and crap like that are still a bit unknown.

An example I am still trying to figure out is how exactly the tune uses the VNT control maps. There are 3 of them and there is no clear indication(to me) when each one is used and why. You can edit the boost map and request more boost, but if the VNT maps dont allow enough vane angle then you wont hit that boost. They work together, where basically the Boost map is the target and the VNT map is the upper limit of % duty cycle the ECU can apply to the VNT control system. If you exceed the boost target then the VNT vanes will back off and lower boost, but if you are below the boost target and the ECU is already applying the maximum duty cycle allowed for that RPM/Throttle, then it wont increase the vane angle to hit the boost. If I wanna increase boost at 3600 RPM, which VNT map do I change and if I change all 3, what are the side effects? I dont fully know. It may select the map based on coolant temp, air temp, ambient air pressure, etc etc etc.

I choose to go with a GDE tune even after the whole MPPS self tuning thing was discovered, and even though I am pretty proficient at tuning now, I still would buy a GDE tune if I had to do it all over again.