trashpanda-org/Qwen2.5-32B-Marigold-v0

Updated 6 days ago • 217 • 3

Note Good char/scenario portrayal and adherence. Speech patterns and formatting is followed well if you use chatml, while being slightly cursed, dramatic and in some cases refreshing when used with Mistral V7. Some testers report user impersonation. Testers also report good resulting prose from having thinking enabled. Not as varied in terms of rerolls compared to Mullein v1.

trashpanda-org/Qwen2.5-32B-Marigold-v0-exp

Updated 7 days ago • 238

Note Second place in our v0 three-model test. Still distinct from v0 with mixed reactions from testers, a little inconsistent with how it's unhinged sometimes and tame other times so if that's what you're looking for, consider trying this out.