---
tags:
- not-for-all-audiences
- nsfw
license: other
language:
- en
---

[EXL2](https://github.com/turboderp/exllamav2/tree/master#exllamav2) Quantization of [Undi95's's MLewd-ReMM-L2-Chat-20B](https://huggingface.co/Undi95/MLewd-ReMM-L2-Chat-20B).


## Model details

Quantized at 3.18bpw with hb 6. Can run full 4K context on 16GB VRAM

Perplexity:

Base = 6.5820

3.18 h6 = 6.6928

Dataset = [wikitext](https://huggingface.co/datasets/wikitext/resolve/refs%2Fconvert%2Fparquet/wikitext-2-v1/test/0000.parquet)

## Prompt Format

```
Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{prompt}

### Response:

```