File size: 826 Bytes
cb69c4c
 
 
 
 
 
 
 
 
 
 
040f4a6
cb69c4c
040f4a6
76153f9
cb69c4c
040f4a6
cb69c4c
040f4a6
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
---
language:
- en
license: apache-2.0
library_name: transformers
tags:
- mistral
- unsloth
- transformers
---

# I quanted this from the Unsloth upload for Mistral Nemo Instruct. 

[You can find the link here](https://huggingface.co/unsloth/Mistral-Nemo-Instruct-2407)
[This is for the base Mistral Nemo Instruct Model](https://huggingface.co/nvidia/Mistral-NeMo-12B-Instruct)

EXL2 quanting seemed to work. I ran a few tests on it and it seemed to have zero issues generating text up to 32k context size. I did not try higher than that, but uploading so folks can start testing this. Pleasantly surprised for a roleplay capacity as it seemed to latch onto character traits very well.

[8BPW](https://huggingface.co/Statuo/Mistral_Nemo_Instruct_EXL2_8bpw)
[6BPW](https://huggingface.co/Statuo/Mistral_Nemo_Instruct_EXL2_6bpw)