metadata
license: other
license_name: mrl
license_link: https://mistral.ai/licenses/MRL-0.1.md
base_model:
- mistralai/Mistral-Large-Instruct-2407
base_model_relation: quantized
tags:
- VPTQ
- Quantized
- Quantization
Disclaimer:
The model is reproduced based on the paper VPTQ: Extreme Low-bit Vector Post-Training Quantization for Large Language Models github and arXiv
The model itself is sourced from a community release.
It is intended only for experimental purposes.
Users are responsible for any consequences arising from the use of this model.
Note:
The PPL test results are for reference only and were collected using GPTQ testing script.
{
"ctx_2048": {
"wikitext2": 16.465784072875977,
"c4": 23.362998962402344,
"c4-new": 28.253490447998047
},
"ctx_4096": {
"wikitext2": 14.906161308288574,
"c4": 22.091983795166016,
"c4-new": 26.17787742614746
},
"ctx_8192": {
"wikitext2": 14.027444839477539,
"c4": 15.35131549835205,
"c4-new": 25.94689178466797
}
}