Recent posts

#21

Off-Topic Community / Re: The YouTube |OT| Place for...

Last post by Legend - Mar 14, 2026, 10:19 PM

Is that Canadian humor? 

Not my cup of tea 




Laughed at these recently:

#22

Off-Topic Community / Re: AI General Thread

Last post by Legend - Mar 14, 2026, 09:16 PM

LM Studio has so many options you can mess with.

You can even set how the context window is managed (truncate middle, rolling). 

Is there a way to run two models in parallel so that the next token can be sampled using both their outputs? Like averaged or multiplied or whatever?

#23

Off-Topic Community / The-Pi-guy is the member of th...

Last post by Legend - Mar 14, 2026, 09:05 PM

Happy Pi day!

#24

Off-Topic Community / Re: AI General Thread

Last post by the-pi-guy - Mar 14, 2026, 08:45 PM

LM Studio has so many options you can mess with. 

You can even set how the context window is managed (truncate middle, rolling).  

#25

Off-Topic Community / Re: AI General Thread

Last post by the-pi-guy - Mar 14, 2026, 06:46 PM

Different sampling method/temperature?

Most of the apps let you change the temperature, top K, top P. 

I'm leaning towards formatting being different, which would be harder to fix as an end user.

#26

Gaming Community / Re: Xbox | Zenimax Development...

Last post by the-pi-guy - Mar 14, 2026, 06:44 PM

Any word on game updates?

https://insider-gaming.com/starfield-future-unveiled-next-week-bethesda/


But apparently Todd said the game isn't going to be wildly different. Not like FF XIV getting a crazy relaunch.

#27

Off-Topic Community / Re: AI General Thread

Last post by Legend - Mar 14, 2026, 05:52 PM

For some reason, running local models feels goofier than it should be. And I'm not sure why. Like for some reason, prompt adherence is worse when I'm using different Android app.

Not sure if they're formatting the requests differently, or passing different default values for the model.

Different sampling method/temperature?

#28

Off-Topic Community / Re: AI General Thread

Last post by the-pi-guy - Mar 14, 2026, 05:35 PM

For some reason, running local models feels goofier than it should be. And I'm not sure why. Like for some reason, prompt adherence is worse when I'm using different Android app.

Not sure if they're formatting the requests differently, or passing different default values for the model.

#29

Off-Topic Community / Re: AI General Thread

Last post by the-pi-guy - Mar 14, 2026, 05:21 AM

AI can be so junky. 

Local model is suddenly not working well. Lots of looping.

I remember this happened before, and I had to downgrade the nvidia drivers. 

But it was not working this time. 

Tried searching with Copilot and it got to a point where it just kept telling me that the model can't run on my computer. Find a smaller model. 

#30

Forum Community / Re: How is everyone doing? Feb...

Last post by Legend - Mar 14, 2026, 03:44 AM

My parents didn't yell at me, yay!!!

Also neither of them had any clue. Nothing. 🤯