---
license: apache-2.0
language:
- en
tags:
- nsfw
- not-for-all-audiences
- roleplay
---
## InfinityKuno-2x7B

![InfinityKuno-2x7B](https://cdn.discordapp.com/attachments/843160171676565508/1219033838454313091/00069-4195457282.jpeg?ex=6609d4bb&is=65f75fbb&hm=4ea1892b3bf2b08040fd84b569ad9f6d4497f6d3d9626d427cb72f229b0218fa&)

GGUF-Imatrix quantizations of [InfinityKuno-2x7B](https://huggingface.co/R136a1/InfinityKuno-2x7B)

Experimental model from [Endevor/InfinityRP-v1-7B](https://huggingface.co/Endevor/InfinityRP-v1-7B) and [SanjiWatsuki/Kunoichi-DPO-v2-7B](https://huggingface.co/SanjiWatsuki/Kunoichi-DPO-v2-7B) models. Merged to MoE model with 2x7B parameters.

## Perplexity

Using llama.cpp/perplexity with private roleplay dataset.

| Format | PPL |
| --- | --- |
| FP16 | 3.2686 +/- 0.12496 |
| Q8_0 | 3.2738 +/- 0.12570 |
| Q5_K_M | 3.2589 +/- 0.12430 |
| IQ4_NL | 3.2689 +/- 0.12487 |
| IQ3_M | 3.3097 +/- 0.12233 |
| IQ2_M | 3.4658 +/- 0.13077 |

### Prompt format: 
Alpaca, Extended Alpaca, Roleplay-Alpaca. (Use any Alpaca based prompt formatting and you should be fine.)

Switch: [FP16](https://huggingface.co/R136a1/InfinityKuno-2x7B) - [GGUF](https://huggingface.co/R136a1/InfinityKuno-2x7B-GGUF)