---
license: cc-by-nc-4.0
tags:
- not-for-all-audiences
- nsfw
---
## Description

Exllama 2 quant of [NeverSleep/Echidna-13b-v0.3](https://huggingface.co/NeverSleep/Echidna-13b-v0.3)

3 BPW, Head bit set to 8

## VRAM
My VRAM usage with 13B models are:
| Bits per weight  | Context | VRAM  |
|--|--|--|
| 8bpw | 8k | 22gb | 
| 8bpw | 4k | 19gb | 
| 6bpw | 8k | 19gb | 
| 6bpw | 4k | 16gb | 
| 4bpw | 8k | 16gb | 
| 4bpw | 4k | 13gb | 
| 3bpw | 8k | 15gb | 
| 3bpw | 4k | 12gb | 
I have rounded up, these arent exact numbers, this is also on a windows machine, they should be slightly lower on linux.

## Prompt template: Alpaca

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

### Instruction:
{prompt}

### Response:

```