Miquella 120B
Model has been remade with the fixed dequantization of miqu.
This is a merge of pre-trained language models created using mergekit. An attempt at re-creating goliath-120b using the new miqu-1-70b model instead of Xwin.
The merge ratios are the same as goliath, only that Xwin is swapped with miqu.
Models Merged
The following models were included in the merge:
- Downloads last month
- 4
Inference API (serverless) is not available, repository is disabled.