DoppelReflEx/MN-12B-FoxFrame-Miyuri

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:12BQuant:FP8Ctx Length:32kPublished:Feb 8, 2025License:cc-by-nc-4.0Architecture:Transformer0.0K Open Weights Warm

DoppelReflEx/MN-12B-FoxFrame-Miyuri is a 12 billion parameter merged language model, part of the FoxFrame series, created by DoppelReflEx. This model is specifically optimized for roleplay (RP) and erotic roleplay (ERP) scenarios, demonstrating an energetic, gyaru-like character in testing. It is built upon a base of IntervitensInc/Mistral-Nemo-Base-2407-chatml and utilizes a dare_ties merge method combining cgato/Nemo-12b-Humanize-KTO-Experimental-Latest, DoppelReflEx/MN-12B-Mimicore-GreenSnake, and MarinaraSpaghetti/NemoMix-Unleashed-12B.

Loading preview...

DoppelReflEx/MN-12B-FoxFrame-Miyuri Overview

DoppelReflEx/MN-12B-FoxFrame-Miyuri is a 12 billion parameter merged language model, part of the FoxFrame series, developed by DoppelReflEx. This model is specifically designed and optimized for roleplay (RP) and erotic roleplay (ERP) applications. It is noted for generating an "energetic, gyaru-like" character in testing scenarios.

Key Characteristics

  • Merge Series: Part of the FoxFrame series, with related models like Yukina and Shinori.
  • Merge Method: Utilizes a dare_ties merge method.
  • Base Model: Built upon IntervitensInc/Mistral-Nemo-Base-2407-chatml.
  • Component Models: Merges cgato/Nemo-12b-Humanize-KTO-Experimental-Latest, DoppelReflEx/MN-12B-Mimicore-GreenSnake, and MarinaraSpaghetti/NemoMix-Unleashed-12B with specific density and weight parameters.
  • Chat Template: Employs the ChatML format.

Usage Notes

Users should be aware that the cgato/Nemo-12b-Humanize-KTO-Experimental-Latest component may occasionally introduce the <|im_end|> token, requiring user intervention (writing more or rerolling the message) to continue generation.

Finetunes
1 models