File size: 1,009 Bytes
881b826
 
 
 
 
 
 
 
 
 
 
ada67e2
 
 
 
 
881b826
 
 
ada67e2
881b826
ada67e2
881b826
ada67e2
881b826
ada67e2
881b826
ada67e2
881b826
df92f1c
881b826
ada67e2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
---
base_model:
- NeverSleep/Lumimaid-v0.2-8B
- vicgalle/Roleplay-Hermes-3-Llama-3.1-8B
- ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.2
- aifeifei798/DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored
- OddTheGreat/Rrat_8B_v.2
library_name: transformers
tags:
- mergekit
- merge
- Roleplay
- Creative
language:
- en
- ru
---
# merge

This is a merge of pre-trained language models

Goal of this merge is to create good 8B roleplay model that will be cpable to RU, for people with low VRAM/RAM.

Tested on 800 replies, on various char cards, from very easy to hard ones.

This model is good on english RP/ERP, and acceptable on russian RP/ERP, it follow instructions, and enough creative.

While not properly tested, assistant is not bad too, both on ru and en.

What about template? Honestly, i don't know. it somehow works fine with LLama 3 instruct, L3 instruct names, L2 Chat, default from silly tavern, and even with ChatML. I liked modded alpaca and default.

Of course, this is 8B, so don't expect very much from it.