Image-Text-to-Text
GGUF
llama.cpp
qwen3_8
abliterated
uncensored
unsloth
imatrix
dynamic-quant
rco
mixed-precision
mtp
Instructions to use ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-Dynamic-MTP-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Local Apps Settings
- Unsloth Desktop
Request for ukisai/Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF uncensored model
#3
by Saddlebag4502 - opened
Thanks for the good work! Any chance you could apply Orca to:
https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF
Smaller footprint, less thinking, uncensored with best method out there
yes please @ajgazin
@Saddlebag4502 @forceware680 RCO Quants here, all four sizes: https://huggingface.co/ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-Dynamic-MTP-GGUF/tree/main/RCO
Note that ukisai also ran an extra fine-tuning pass ( "GSQ refinement" ) for a small quality increase. I didn't do that step -- it needs more compute than I have and code that isn't public. So these are "RCO" files, not "GSQ-RCO".
I'll do a full KLD pass tomorrow or so to see how they stack up to the UD quants.