mradermacher

/

model_requests

like 197

yay, nice sheeps I hope. Any new sota you recommend to try? =)

It's queued! =)

You can check for progress at http://hf.tst.eu/status.html or regularly check the model
summary page at https://hf.tst.eu/model#Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-GGUF
https://hf.tst.eu/model#Qwen3.6-27B-V1.1-FF711-Darker-Hero-GAIN-H2.0-GGUF
for quants to appear.

Thank you!

This is SOTA (30 days old only) - exceeds performance on all Qwen 3.6 27B base and fine tunes:

2100+ likes, 3 million + downloads, 3rd party verified benches and performance, 61+ quant repos.

https://huggingface.co/DavidAU/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-NEO-MAX-MTP-GGUF

We plan to do the same with Qwen 3.8 27B ; the training pipeline takes 7-10 days min however.
That being said, the detail/work shows in FF711.

Cold Fusion/GAIN is a new training method born from R/D with FF711 pipeline devel ;

Thanks again for all you do.

oh nice, good luck then =)