metadata

license: apache-2.0
pipeline_tag: text-generation
tags:
  - gguf
  - mergekit
  - merge
  - mistral
  - not-for-all-audiences
  - nsfw
  - rp
  - roleplay
language:
  - en

This model is recommended for RP, but you can use it as assistant as well. Please, give it a try.

Version v3 is here.

Prompt Format:

Extended Alpaca Format As for exemple from lemonilia/LimaRP-Mistral-7B-v0.1. Use ### Response: (length = huge) for exemple, to increase length.

Key features (What I observed):

Merely good reasoning.
Follows, understand and remember character card well even at 9k~ context. I haven't tested more than 9k yet.
Good for multi-turn.
Good for creative writing, creating concise stories/texts without repetitiveness or hallucinations on my tests.
Follows the character personality well. I tested with a lot and it gave good results.
If you like * asterisks rather "quotes" or vice versa, this model is for you. It does not mistakenly use quotes to describe actions.

Edit: I tested with quotes and it works very well for dialogs.