Man I am excited about getting a English release of #ToHeart I just wish I did not hate the "new style" compared to how it use to look. Maybe I am just a boomer
https://store.steampowered.com/app/3380520/ToHeart/
https://youtu.be/4dBAWcYGBIg

https://store.steampowered.com/app/3380520/ToHeart/
https://youtu.be/4dBAWcYGBIg

@sun
Though in reality this is just a lame python GUI front end for interacting with a Qwen2.5 vision model over an api with folder watching. But I just like the concept and thought it might be fun to try and play Doki Doki Pretty League.
Though in reality this is just a lame python GUI front end for interacting with a Qwen2.5 vision model over an api with folder watching. But I just like the concept and thought it might be fun to try and play Doki Doki Pretty League.
@sun
Planned on releasing what I got this weekend. Wanted to work through some different prompts/workflows make sure I am not missing anything.
Planned on releasing what I got this weekend. Wanted to work through some different prompts/workflows make sure I am not missing anything.
@queenofhatred @bl00d Executive dysfunction is the worst.
@sun This is good to know I have not tested much NES-era games. Curious how well the model picks up the font. Thanks for looking out!
@lain In your opinion, if Xenoblade 1 bores me to tears can I enjoy X and later games?
@CwalkPinoy As a JF user I can confirm. So many times I plug in a USB drive to the TVs that support it.
I want to do a good write up in my README for PixelPolygot as one of my last touches but I need the damn #rocm fork of #KoboldCpp to update so I can do some more testing with Qwen2.5-VL locally. Like it works with vulkan on the main branch but way slower than rocm. 

Don't really get excited for #Miku or #Vocaloid stuff, but I really wanna check this out
https://gmpj.bn-ent.net/ja
https://gmpj.bn-ent.net/ja

@quad A tool I am working on that uses a low powered LLM (running locally or on a api) to extract the Japanese text from a image and translate it. The main feature is you can set a directory for it to watch and anytime a new image enters it auto sends it off to the LLM to get processed. So if you are playing something in #RetroArch or another emulator you have this side by side and press the screenshot button when you need a translation. I think I posted a video of a earlier version a week or so ago here:
https://melonbread.dev/notice/AscmWuE6CCbWpvsSZM
Still working some things but will share the repo soon once I work a couple more things out.
https://melonbread.dev/notice/AscmWuE6CCbWpvsSZM
Still working some things but will share the repo soon once I work a couple more things out.
Was able to implement it but need to work with some of the prompts more. but can easily switch between prompts based on the situation.


I think the trick might be to have a DropBox UI element that lets you load different prompts based on how the game is set-up. Like this prompt might be nice for a game with menus and such but if you are just playing a VN like game you would just want dialogue. Could even make prompts for specific games if the text layout is weird. 

I like this layout of text a bit better. Though I think I need to tweak a bit so notes are not so aggressive when not needed.
(Game is Medarots 9)
(Game is Medarots 9)

Surprised with how much text it was able to get out of this one image. I might have to work with the prompt for better handling of a bunch different strings of text in a table or something. Not that I think anyone would ever use this tool for something like Dragon Quest X.