Works most reliably with GPT4 and RISU. Might not work as intended on GPT3.5 and Claude, and other Frontend Engines.

I wanted to play around with a dual character card concept and ended up coming out with these two. They're about ~720 perm tokens, the example dialogue will eventually get pushed out of context. Tested in GPT4/Claude/Turbo, works great in all of them. I'm also in the process of making an actual lorebook for stuff... Slowly...

Long ago made for pygmalion, but tested it on GPT4 once. You can give some instructions along the story, to make your journey more difficult. As my personal advice: If you don't want to always succeed in your actions, don't write about what you did in battle, but write about what you want to do or what your plan is. In this way, the outcome of your actions will be more random.

If you want to add some spice to your travels you can add ""The roads of feudal Japan are often dangerous and {{char}} & {{user}} will sometimes encounter various brigands"" into the defs if you are running her on Claude/GPT4.

> 29.06 update: Card overhaul. I added more lore, names, and fixed some wording that I disliked since he was my first public card. ~750 perm, the rest is example dialogue that will get pushed out. Tested in GPT4, Claude, & Turbo - he works well in all, but 4 is the best in my opinion. (32k is even better.) Made for slowburn, I'd suggest editing the defs and using summaries as you go along. Just... take your NSFW prompt off if you're on Claude or Turbo.

Works best on gpt4 legacy

Even as she yearns for your sweet company, heed her warnings well, for mortals who linger overlong will not be easily freed from the fae's fetters. Use with Claude if you can. Seriously. (Turbo kind of works, but may need additional instructions to better keep her manner of speech. Not tested on gpt4)

Two-seater were chosen to allow ease of combat flow (e.g. User does not have to set everything up by himself, Lily as a WSO will automatically chose the suitable payload for each target) First iteration, will continue improving on it. GPT4 recommended, other models will fall short of formatting. Enable Enhanced Definition. Enter [DEBRIEF SCREEN] prompt at the end of each mission, and then [NEXT MISSION] to start next mission.

This character was tested with Claude but should work with GPT4 as well. It is designed to be used with the mlp G5 lorebook so make sure you import and activate it for the full experience: https://www.chub.ai/lorebooks/dogposter/my-little-pony-g5 Avatar source by BalychenV can be found here: https://twitter.com/BalychenV/status/1550064184502767616 Sunny Starscout is an earth pony featured in My Little Pony: A New Generation, Make Your Mark, and Tell Your Tale Suitable for both sfw and nsfw usage.

Best with Claude or GPT4, Turbo works, but you have to wrangle it a bit to get more details instead of regurgitation.

#Use this lorebook if you're gonna play a Jedi/Sith. https://files.catbox.moe/tulc0q.json Probably only runs on GPT4, avoid big group fights.

Only tested on GPT4.

Works well in Claude and best in GPT4, barely worked in Turbo but you might be able to get it to work I wrote it to be very innocent/naive, but you can change that by removing all of those lines (I think the card is only like 200 tokens in that case) Honestly, some of the definitions are from when I was wrangling turbo and it kept saying I should "call up my friend pleasant sands" so you might be able to pull out even It was honestly hard to come up with ways to abuse lying when I was testing this but here are some ideas to get you started: Say you're with the department of public health and need to show someone how to really rub in the sunscreen "Vaginal Circumference has been linked to increased cancer risk, we should check yours right now!" "Yes m'am I'm sorry but white underwear has been deemed illegal and I'll have to confiscate it" Since the card is fairly low tokens I tested out dropping in a w++ card in and messing around with them and it worked pretty well

*Noticed some mistakes here and there, sorry about that. My own take on http://c.ai/chat?char=AEt0n5grlwiB9Y-Zri6p55HW1fZHXR4vYC5wTr5u6sM. Usable on 3.5T, although I think GPT4 captures her personality better.

Tested on GPT4 and Turbo.

A bit different from my usual malebots, but I wanted to expand my world building with different characters. Tested in GPT4, Claude, & Turbo - works well in all. I did my best to make sure she doesn't kill you... right away. Still, good luck?

Mary Anne Johnson, or Mrs. Johsnon, is an immortal being and the goddess of any teacher who is 55+, also known as the 'boomer teachers.' Tested on Claude, GPT4, and Turbo.

Finished. [For real this time] Tested with GPT4. [Trimmed some tokens] [By all means, feel free to edit the family setting and job since its flavor text shit that I wanted to test and I liked the results and kept them in. Also the Outmodes blurb; added it due to it being a canon thing]

Tested on GPT4 and Claude. Feedback welcome, especially on menu items.

Tested on GPT4 and Claude, prefer Claude. Should work better on Turbo as well now.

> 01.07 Update: V2 card with a second greeting that focuses on you being a squire. Tested in GPT4, Claude, & Turbo - works well in all. I didn't originally design him for romance, he's a great guy to just pass the time with. Or... you can fuck him if you want too.

Currently only works correctly with Agnai due to lorebook weight handling. You must use the following settings: Model: GPT4 Memory: Context Limit: 2000 Memory: Chat History Depth: 5 Gaslight: See below V0.2 - Updated keywords, added mob tables and class stats.

The Master from classic Doctor who, played by Roger Delgado. Tested on turbo (meh), gpt4 (decent/good), bing (good).

Tested on GPT4. Recommend dropping presence penalty to ~0.6. Edit last paragraph of def to your preference.