brucethemoose

> > For what its worth, the issue went away for me running sddm-git (from the CachyOS Arch repo). > > I'm not sure what you mean. Did you uninstall...

Could more "advanced" merging be supported by similar pipelines? Namely this, which produces better results when merging models with different "base" SD ancestors: https://github.com/ogkalu2/Merge-Stable-Diffusion-models-without-distortion And this, which allows for specifying...

Well for starters the install for the new branch doesn't quite work? I had to manually add the scripts and merge_methods folders into pip's install directory. Mergekit doesn't like the...

> That looks like a tokenizer mismatch issue to me. Did you maybe copy in the tokenizer for Tess-M-v1.2? The added tokens not present in the base model could cause...

Yeah it works with the base model tokenizer, thanks. In fact, a few responses from the merge model seem pretty smart.

Any positive results from parameter tweaking yet? Also, is there a particular reason not to go higher density? Should't values above 0.5 "preserve" more of the finetuning from the models?

On second thought, the internlm deployment framework is pretty interesting, albeit not quite right for my personal and work use cases.

Yeah, optwritemotion used to straight up crash some modpacks in my testing. Are you testing `WriteableCodeCache` as well?

> Yes I use Ah now that's interesting. WriteableCodeCache improved chunkgen times in my 22.1.0 benches (which is unusual, and most flags dont improve chunkgen throughput like that), but it...

> > > Yes I use > > > > > > Ah now that's interesting. WriteableCodeCache improved chunkgen times in my 22.1.0 benches (which is unusual, and most flags...