Three-model dare_ties merge
According to the model card, TroyDoesAI/BlackSheep-24B, TheDrummer/Cydonia-24B-v2, and PocketDoc/Dans-PersonalityEngine-V1.2.0-24b were merged with dare_ties.
Open Source Model Profile · DoppelReflEx
MiniusLight-24B-v2.1 is a 23.57B-parameter Mistral merge from DoppelReflEx. According to the model card, it uses dare_ties mixing with ChatML formatting.
MiniusLight-24B-v2.1 is published by DoppelReflEx as a text-generation model. The captured configuration identifies MistralForCausalLM with model type mistral, and Safetensors metadata reports 23,572,403,200 parameters. According to the model card, it is a dare_ties merge of three 24B sources with a documented ChatML template.
According to the model card, TroyDoesAI/BlackSheep-24B, TheDrummer/Cydonia-24B-v2, and PocketDoc/Dans-PersonalityEngine-V1.2.0-24b were merged with dare_ties.
According to the model card, ChatML is named as the chat template, with Mistral V7 as an alternative.
Captured Safetensors metadata reports 23,572,403,200 parameters for a MistralForCausalLM record with model type mistral.
Source: DoppelReflEx/MiniusLight-24B-v2.1
Captured: Unknown. Processed: 2026-09-07T19:35:04.738040+00:00.
MiniusLight-24B-v2.1 12B - 24B-v1 - 24B-v1.01 - 24B-v2 - 24B-v2.1 Origin Content (Click Here) What is this? A merge of most uncensored model TroyDoesAI/BlackSheep-24B and recipe of MiniusLight-24B: TheDrummer/Cydonia-24B-v2 and PocketDoc/Dans-PersonalityEngine-V1.2.0-24b. Another version of v2, but far better than it. Vivid writing styles, and talk back to me, sometimes hard to control it. (Maybe just because my character card) Best model of the series (for me). :) PS: Highest NatInt for 24B model in UGI leaderboard (1st May 2025) GGUF (Thank mradermacher and his team so much (nicoboss too)) Static - iMatrix Other information Chat T…
F001F002F003F004F005F006F008F009F011F012F013F014