Ren'Py Lab Rats 2 - Reformulate [v2025 4.0.37]

[Magister Crudi]

Guys, the GenAI mod only works with AUTOMATIC1111 version of Stable Diffusion.

And correct me if I'm wrong, but these versions only work with the 'market' version of LR2?

Whenever I try to use them with '2025.07-beta' they crash as soon as I select the 'GenAI Config' screen.

A shame, as the code looks more elegant than AnotherMike's and I'd use it as the basis of my own tweaks if it worked with both versions. Not having access to the main LR2 beta source means I'd be working in the dark if I tried to fix the crash.

 

[NaughtyAnon]

And correct me if I'm wrong, but these versions only work with the 'market' version of LR2?

Whenever I try to use them with '2025.07-beta' they crash as soon as I select the 'GenAI Config' screen.

A shame, as the code looks more elegant than AnotherMike's and I'd use it as the basis of my own tweaks if it worked with both versions. Not having access to the main LR2 beta source means I'd be working in the dark if I tried to fix the crash.

I have been running the last release by

You must be registered to see the links

on 2025.07-beta + VTMod4.042 without any crashes or issues. The other version, while more detailed, seemed to crash the system far more or just not generate well in testing.

Edit for details: I am running WebUI Forge with CUDA 12.1 and Pytorch 2.3.1 on an RTX4070TI for generation and ADetailer.

 

[asdasd5565]

Using the latest modded version 2025.07-betaVTMod4.0.37. How come I cannot do anything besides groping during the night scenes? This is day 0 and the first Night event after work. Is oral behind some stat requirement or perk now?

 

[Selek]

My rig's fairly beefy but not as much as yours - i5-14600k processor, 32GB RAM, RTX 4070 Super with 12GB - but I'm still getting 20-25 steps in around 5 secs with most model-sampler combos and ADetailer and Upscaling turned off. Have you got xformers installed and activated in COMMANDLINE_ARGS?

Ah, I did not have xformers installed. (I'm new to Stable Diffusion.) I just tried adding them, and I'm not sure I'm seeing a difference, so I'm not sure I installed correctly. I added --xformers to the command line (in webui-user.bat), and when Stable Diffusion loaded, the command-line window indeed reported that it was installing xformers. (I used to get a 'no xformers found' warning.) But at 20 steps, it takes me more like 10 seconds, and 10 steps still takes about 5-6 seconds. I don't have ADetailer, and I have Upscaling turned off. I'm using the DreamShaperXL model right now. (It does seem generally slower than Gonzales, say.)

How can I check that I've installed xformers correctly?

Edit: I removed --medvram from my arguments list, and my performance improved dramatically. I have 24GB of VRAM, so I didn't need that argument. I can now do 20-25 steps in about five seconds, using my slowest model checkpoint (the DreamShaper XL). But I actually think I prefer the DreamShaper with fewer steps. With more steps, I start seeing more blemishes and such on legs and arms, which may be more realistic, but it's less appealing visually. Anyway, I'm pretty happy with performance now.

 

[vanhelz]

I mentioned this on the original thread. "The best way to improve the game to a whole new level instead of these micro changes is to add real world consequences to the game. You're selling a lot of these serums? To whom? Wouldn't the game be a lot more amazing, if you see random women on the street having suggestibility coz they took the serum? Or their body changes, or sluttiness increases? Having suspicion levels so you have to start off small then as people acclimate going to next tier. "

 

[AECOBoiE]

I find it's all over the shop depending on what models you use. A good example is that few models know what 'Capris' are and so render the girls wearing them as bottomless.

I also find most models will spit the dummy beyond a certain level of prompt complexity, giving you monsters or splatterings of colour only Jackson Pollock would see as women.

I've now tried to customise AnotherMike's GenAI to several different models. StableYogi ones tend to give the best sex positions (I'd recommend realismByStableYogi_ponyV3VAE and his Porncraft models) but tend to fall short on clothes, appearances of women and backgrounds. Cyberrealistic ones are fairly good all round but are more likely to generate monsters. Both have their own sets of positive and negative prompt files (textual inversions) but they don't fix everything.

ADetailer can fix a few minor glitches (e.g. hands) and can help a lot with faces - especially when they're upside down in the piledriver position - but I'm finding it more than doubles image generation time, so when actually playing the game instead of messing with images you end up clicking through most of them and missing it. Ten seconds per image is about as far as my patience stretches and I like to keep it close to five seconds.

When I get a bit of time I'm going to start training a model to specific prompts I'll use to customise the GenAI code, but that's a fairly long term project.

Here's the tweak of AnotherMike's prompt builder I'm currently trialling with

You must be registered to see the links

and its

You must be registered to see the links

and

You must be registered to see the links

prompt embeddings. Right now I'm trying to make 'kneeling oral' work consistently. I'm using the 'Euler a' sampler, with 25 steps, CFG 9 (!) and 0.35 denoising with both img2img and basic prompts turned off. Mostly I leave upscaling and ADetailer off but sometimes switch them on for sex scenes.

Excuse Me, Which GenAI Version are you using?. Im getting a crash with your Prompt builder

 

[Magister Crudi]

Excuse Me, Which GenAI Version are you using?. Im getting a crash with your Prompt builder

It's one of AnotherMike's. Here's the whole thing.
Don't forget to either delete the .rpyc files from the GenAI folder or force a full rebuild with shift-R.

Edit: 11-Oct-2025. GenAI.zip replaced with V0.01 alpha

 

[Magister Crudi]

How can I check that I've installed xformers correctly?

If you're not generating errors or warnings in the startup I think it's a safe bet they're installed correctly.

Dunno what the issue is here. Presumably something about your setup is slowing things down.
Right now I'm using cyberrealisticPony_v130 with 25 steps and consistently getting images in 5-6 seconds.

Have you got the latest Nvidia driver?
What about CUDA? I'm running 12.8.

What Python version are you using?
I'm using 3.10.6 because people report problems using SD with later ones. (LR2 itself uses 3.9 libraries)

Edit: I just had the experience of switching ADetailer off from the config menu only to find the image generation still took as much time as when it was on. After switching it on and off a few more times it started working properly again. Dunno if you're having a similar problem.

 

[Magister Crudi]

I have been running the last release by

You must be registered to see the links

on 2025.07-beta + VTMod4.042 without any crashes or issues. The other version, while more detailed, seemed to crash the system far more or just not generate well in testing.

Thanks for that.

The good news is it runs without crashing.

The bad news is the images generated don't match what's happening in the game. As the UI doesn't display generated prompts like AnotherMike's it's hard for me to see what's going wrong.

Later I'll take a closer look, but if you've got a tip I'd appreciate it.

 

[Magister Crudi]

So I've run CyberRealistic XL V7.0 and never got anything this... interesting.

I only got it once myself, but it was odd enough to be worth mentioning. It's not an artifact so must be coming from training images of people with stuff like that in their face, so I wondered if there's a whole realm of perversion that's passed me by.

BTW, your XL V7.0 characters are close to perfect matches to my Pony V13.0 ones. Ashley and Stephanie look the same and your Ellie looks like a slightly chubby sister of my Ellie.

 

[NaughtyAnon]

Thanks for that.

The good news is it runs without crashing.

The bad news is the images generated don't match what's happening in the game. As the UI doesn't display generated prompts like AnotherMike's it's hard for me to see what's going wrong.

Later I'll take a closer look, but if you've got a tip I'd appreciate it.

The biggest thing about the previous versions that AnotherMikes was built off of is that it has the sampler setup but nothing to adjust the scheduler. Previously I hard coded my own but the later version by Maswoo he changed it to use a prompt system instead. While more work I can setup a single file to drive the specific model, sampler, and scheduler, while also customizing the prompt generation to best fit the model. While not perfect I can fix the really bad ones not matching the game pretty easy. Hope this helps and happy modding!

 

[NaughtyAnon]

I only got it once myself, but it was odd enough to be worth mentioning. It's not an artifact so must be coming from training images of people with stuff like that in their face, so I wondered if there's a whole realm of perversion that's passed me by.

BTW, your XL V7.0 characters are close to perfect matches to my Pony V13.0 ones. Ashley and Stephanie look the same and your Ellie looks like a slightly chubby sister of my Ellie.

Its fun to see how close the models get especially when they are trained off similar base models. One test I ran which was crazy fun was throwing a Retro LoRA into the mix.

 

[slobber]

I mentioned this on the original thread. "The best way to improve the game to a whole new level instead of these micro changes is to add real world consequences to the game. You're selling a lot of these serums? To whom? Wouldn't the game be a lot more amazing, if you see random women on the street having suggestibility coz they took the serum? Or their body changes, or sluttiness increases? Having suspicion levels so you have to start off small then as people acclimate going to next tier. "

that sounds more like Course of Temptation type of setting.. which would be a bit tougher to do since Lab Rats is more involved graphically and feels more clunky i think. CoT is more text based so it has the room for it.

edit: i'm happy with the direction LR is going and i'm happy when there's more story for the main love interests

 

[vanhelz]

that sounds more like Course of Temptation type of setting.. which would be a bit tougher to do since Lab Rats is more involved graphically and feels more clunky i think. CoT is more text based so it has the room for it.

edit: i'm happy with the direction LR is going and i'm happy when there's more story for the main love interests

what? The mechanisms are already there though? We have suspicion level and others right? Is it difficult for npcs to have random "serum needs" assigned same as they get assigned random traits? This makes some serums easier to use on them while having more difficulty with others? Or for a fraction of the serums sold to be affecting the reality of the town?

Once you reach a certain level of sales or revenue, what's the point of serums being sold other than taking your attention away from using serums on people?

 

[Darkmind]

Guys any tips how to run this on Mac?

There's an old Mac build, but if you want current releases the easiest way is through Crossover. Just set up a standard environment, copy the game in, and run it through that.

The AI mod even will work just fine with StableDiffusion running natively on the Mac, as long as you have things configured correctly to open the ports.

 

[SDN111]

I mentioned this on the original thread. "The best way to improve the game to a whole new level instead of these micro changes is to add real world consequences to the game. You're selling a lot of these serums? To whom? Wouldn't the game be a lot more amazing, if you see random women on the street having suggestibility coz they took the serum? Or their body changes, or sluttiness increases? Having suspicion levels so you have to start off small then as people acclimate going to next tier. "

The problem with that idea is three fold.

1. It breaks the current progress levels in the game. People being effected by the serums without having to meet them will cause general issues of speeding up too greatly the development of the city and employees. Aside from that you have the logical issue of men and women should be effected and that customers also should have been outside of the range of the city.

I am not going to say this is a bad idea. Ideally we could have standard customers who we know are in the city (game lounge [kinda implemented], mom's business, strip club [already implemented], bar, school). Ideally we should have males being able to be effected in more than cutscenes and our beneficial ones. But this flows into our second issue.

2. Implementing this in code. How would it be decided who is effected and keep track of that. The best way to do this would be to have other businesses have inventory. Have logic on which employees recieve it or customers order it and then make custom orders for the businesses. Something like this aside from processing overhead increase for each turn also creates new bottle necks in your manufacturing that would need to be fixed/addressed. It would make a decent difficulty spike but again Ideally male characters should both exist and be effected by this.

3. Performance will suffer keeping track of all of it. The game will noticeably slowdown when too many serums are in effect and/or too many characters exist. I know I have had the game crash from having too many copies of a serum on one person and by having too many clones in the world. This is an optimization problem that would require an entire rework of how everything works. The current method of checking for effects works well and I honestly do not know of a better way do it with the large variety of traits, people and how they can interact. The problem is that this will scale stupidly quickly requiring more resources for each turn. There are only two potential solutions for this. 1. 1 turn serums based solely on location this would then use location data to apply serum enforced changes or 2. Simplify serums so we can track prebuilt effects over time. In theory there is option 3 which would be reducing the number of characters to track but that would go against current game mechanics.

A large amount of the problem is the scale of the gameplay and the limitations built into the original base game.

 

[Selek]

It's one of AnotherMike's. Here's the whole thing.
Don't forget to either delete the .rpyc files from the GenAI folder or force a full rebuild with shift-R.

I see just one .rpy file in the zip file you just posted, and three .py files -- nothing else. Do you mean to delete the .rpy file before putting this folder into the mod directory? Also, do you type shift-R once you're in the game?

Dunno what the issue is here. Presumably something about your setup is slowing things down. ... What version of Python...

Things are running very fast now. I removed the --medvram argument from my batch file, and immediately my performance zoomed. I can do 30 steps in 4 seconds or so. I have a 24G VRAM video card (4090), so I suppose the medvram argument (recommended by Maswoo in his install instructions) was throttling things. As for my Python version, I don't seem to have Python installed at all! At least I find nothing when I type python --version into the Windows cmd prompt. And yet the GenAI mod is running smoothly, so I'm reluctant to mess with things too much now.

Edit: I love the face quality with DreamShaperXL, but the bodies are a little same-y. Busty doesn't look that different from a slight build. And DreamShaper doesn't handle naughty stuff as well as Gonzales (whose faces aren't as nice). I'm tempted to try CyberRealistic XL 7, the model you're all discussing. Is it better? Also, I see a 6G version (pruned model fp16) and a 12G version (fp32). Can I get away with the 6G one?

 

[boringapocalypse]

I have a pretty new rig, 96G RAM, nVidia RTX 4090 with 24G VRAM, but even just 6 steps takes about 5-6 seconds using Dreamshaper. (The dev recommends 4 steps by default.) Is there any downside to increasing the number of steps, other than wait times? Do more steps help prevent things like extra limbs or deformed hands?


I've started leaving Auto Generate on, and I like it -- I sometimes miss new images when I'm playing without it. But you're right, sometimes Auto misses changes in the character, in which case I'm inclined to generate a new image. Especially in sex scenes, heh.

The impact of step # also depends a lot on which sampler setting is being used, Euler alpha basically gets no benefit past 30 steps. Also, excessive step # can actually ADD extra limbs and deformities as the AI continually tries to add shit to what is basically a complete image. Source: experience

 

[Selek]

The impact of step # also depends a lot on which sampler setting is being used, Euler alpha basically gets no benefit past 30 steps. Also, excessive step # can actually ADD extra limbs and deformities as the AI continually tries to add shit to what is basically a complete image. Source: experience

Very useful comments -- thanks! Yeah, I actually prefer DreamShaper with fewer than 20 steps. More steps seems to add more noise and blemishes to legs, in particular, but also to background elements and arms.

 

[Magister Crudi]

Do you mean to delete the .rpy file before putting this folder into the mod directory? Also, do you type shift-R once you're in the game?

I mean the rpycs in the version of GenAI already in your mods directory (if you don't just delete the whole mods\GenAI directory before installing the one I posted and start a new game). But yeah, shift-R when you're in the game will force a rebuild of all the rpyc files from their rpy equivs and is the method I prefer.

Things are running very fast now. I removed the --medvram argument from my batch file, and immediately my performance zoomed.

That explains it. Congrats on the fix.

CyberRealistic XL 7, the model you're all discussing.

I use the 12G

You must be registered to see the links

version because it works well with its matching

You must be registered to see the links

and

You must be registered to see the links

prompt embeddings and can do a good range of sex positions if you twist its arm enough with the right prompts. The images of women are attractive but with little facial variation, including with age. I tend to get Rebeccas who look younger than Gabrielles, etc. It reminds me a bit of the old Archie comics where Veronica and Betty had identical faces and bodies and only differed in hair and clothes. But you can fix some of that with the 'Append' fields in the UI. There's also an intermittent issue with red artifacts - random items of red clothing or sometimes even red skin - appearing on the images but it's not common. I haven't been able to find a negative prompt that eliminates it.

I haven't used XL 7 but it looks very similar from what I've seen posted. Can't speak for the pruned versions.