# bloatmaxx 
-> %grey%by%% %orange%meatrocket%% <-
-> %grey%meatyrocket@proton.me%% <-
-> %grey%discord:%% %orange%meat.rocket%% <-
!!!warning bloated preset... about ~4-7k tokens excluding card + chat

-> [![](https://files.catbox.moe/6ofm7r.png)]() <-
-> [![](https://files.catbox.moe/ld4s9d.png)]() [![](https://files.catbox.moe/op9ail.png)]()  [![](https://files.catbox.moe/9ipp83.png)]() <-

-> %grey%*for Anthropic, OpenAI, & Google LLMs.*%%  <- 

%grey%Latest Downloads%%
[TOC2]

# -> %grey%commonly asked questions%% <-

!!!info look forward to new updates, im always improving these presets from time to time

> wtf is this?
this preset includes prompts that will generate some interactive html panels that is based on the nature of your roleplay
includes a **SillyTavern extension to support HTML button inputs from AI** (for kino text adventures)

> why is it so fucking large? 1k+ tokens is bloat enough
this includes a bunch of fluff for multiple models to work with. 
the instruction set alone for the HTML + options excluding any examples is about 2.4k tokens in total.
was originally for furbo, now includes both claude 3, gemini & mistral.

>how to install?
- **download** the desired preset below
- **import** the preset into your sillytavern's chat completion tab
- **enable** prompts/prefill as needed

>what model do i use?
whichever is the best, smartest (for RP & complex instruction following) model available for you to use

my personal RP ranking between the common models here:
```
in order from best > worst
| = same in ranking

claude 3 opus > claude 3 sonnet | gemini 1.5 pro > claude 2.1 > gpt-4-32k | gemini 1.0 ultra > gpt-4 turbo | mistral-large  > gemini 1.0 | mistral medium

separated by model family:

claude 3 opus > sonnet > claude 2.1 
gemini 1.5 pro > 1.0 ultra
gpt-4-32k-0613 > gpt-4-1106-preview / gpt-4-0125-preview
mistral large > mistral medium
```
originally made for furbo due to its release speed (30+ tk/s) but its trash and slow now. use **gpt-4-32k** if you need to
assistant prefills are at the bottom of the preset

>which model do you recommend?
nowadays i pretty much exclusively use **Opus**, although I do swap to other models, notably, Gemini Ultra/1.5 Pro & Mistral 
(i rarely use gpt-4 nowadays, it is a SOVLLESS model...)
however, i do highly recommend Claude 3 **Sonnet** as it's a pretty fast, cost effective model compared to GPT-4 & Claude 3 Opus, which also beats Claude 2.1 in RP by its own right.

> what are your recommended parameters?
claude 3 opus/sonnet:
```
Input Tokens: 24,000 - 200,000 
Output Tokens: 4096 

Temp: 0.9 - 1.0
Top K: 0 
Top P: 0.75 - 1.0
```
gpt-4 32k/furbo:
```
Input Tokens: 8,192 - 32,000 
Output Tokens: 3200 

Temp: 1.65 - 1.95
Freq. Penalty: 0 
Pres. Penalty: 0.05 - 0.10
Top P: 0.65 - 0.80
```
gemini 1.5 pro/1.0 ultra:
```
Input Tokens: 8,192 - 32,000 
Output Tokens: 2000 

Temp: 0.9 - 1.0
Top K: 0 
Top P: 0.9 - 1.0
```
general:
```
Character Names Behavior: Message Content
Continue Postfix: Space
Squash system messages: ON
Send inline images: ON
```

>how do you get images? how does it work?
enable the prompt and the corresponding prefill toggle (depending on your version of the preset)
it uses an external service ("pollinations.ai") to prompt an image generation via url

- - -
# -> %orange%downloads%% <-
#### -> **[ST extension by horsearmor](https://github.com/horse-armor/st-clickable-inputs)** <-
-> **clickable html elements in ST, for HTML &  text adventure prompts** <-

#### -> [[latest release] opus (emojiless) v1.1a (1/04/24)](https://files.catbox.moe/zgbqa9.json) <-
-> bug fixes, making sure it works smoothly with a majority of cards, trimmed examples heavily <-\
-> removed all mentions of emojis <-

#### -> [opus v1.1 (1/03/24)](https://files.catbox.moe/6iqffp.json) <-
-> v1.1: bug fixes for formatting with the extension, trimmed examples <- 

#### -> [opus v1 (31/03/24)](https://files.catbox.moe/zfqcy3.json) <-
-> streamlined things a bit for opus, updated instructions and prose <-
-> WORKS on **claude 3 opus/sonnet, gemini 1.5 pro / 1.0 ultra & gpt-4-32k & gpt-4-1106** <-

-> [![](https://files.catbox.moe/js9n7v.png)]() [![](https://files.catbox.moe/dvw87c.png)]() <-
-> includes brand new `codeblock` panels that are similar in principle to the HTML panels, but smaller in token output. <-
-> also includes html + text adventure prompts from previous v10<-

#### -> [bloat v10 (31/03/24)](https://files.catbox.moe/mbb9pr.json) <-
-> updated html instructions, added a vision prompt and a brand new text adventure prompt<-

-> [![](https://files.catbox.moe/s1r78z.png)]() <-
-> [![](https://files.catbox.moe/u7vsl5.png)]() <-
-> (just enable the corresponding prompt + prefill) <-

-> [![](https://files.catbox.moe/am18x7.png)]() <-
-> (works well with the ST extension) <-

#### -> [bloat v8 (16/03/24)](https://files.catbox.moe/qs4bcw.json) <-
-> rearranged prompts for opus<-

#### -> [bloat v7 (07/03/24)](https://files.catbox.moe/jlbz2r.json) <-
-> leaked version from sturdy<-

#### -> [bloat v1 (19/11/23)](https://files.catbox.moe/72sn3p.json) <-
-> first version of bloatmaxxing<-

#### -> **[[experimental] claude rp (31/03/24)](https://files.catbox.moe/0t4fwe.json)** <-
-> experimental [autistic] prose, heavy usage of alt characters<-

#### -> **[[experimental] mistral rp (17/03/24)](https://files.catbox.moe/jp17da.json)** <-
-> mistral prompt preset, includes html <-

#### ->**[alternative, lighter (2.4k tokens) version (16/03/24)](https://files.catbox.moe/5iesg6.json)** <-
-> use this if you're a paypig/proxyfag with a quota <-

--- 
-> [![](https://files.catbox.moe/k60sr9.png)]() <-
-> [![](https://files.catbox.moe/a6ikv9.png)]() <-
> (generated with sonnet)
- - -
-> [![](https://files.catbox.moe/fvm4kq.png)]() <-
-> [![](https://files.catbox.moe/wlo8c3.png)]() <-
-> [![](https://files.catbox.moe/kqihb4.png)]() <-

---

special thanks to:
- horsearmor (for the [extension](https://github.com/horse-armor/st-clickable-inputs ), proxy owner) 
- mm (examples, [proxy owner](https://rentry.org/mysteryinfo)) 
- branon (proxy owner, and [for making this card](https://www.chub.ai/characters/Branon/shin-megami-tensei-simulator-v2-0-d1ac08fc))
- XMLK (followed their [guide](https://rentry.org/CharacterProvider) for ideas)
- lurker (huge help with prompting)
- sturdycord