DarkArtsForge/Asmodeus-24B-v3

TEXT GENERATIONConcurrent Unit Cost:2Model Size:24BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 20, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

DarkArtsForge/Asmodeus-24B-v3 is a 24 billion parameter uncensored merge of MistralForCausalLM-based language models, created using mergekit's flux method. This model is specifically designed to generate narratives and roleplay content without refusals, including violent, graphic, and NSFW material. It excels in creative and unrestricted content generation, making it suitable for applications requiring complete freedom in output. The model operates with a 32768 token context length.

Loading preview...

DarkArtsForge/Asmodeus-24B-v3: Uncensored Content Generation

Asmodeus-24B-v3 is a 24 billion parameter language model developed by DarkArtsForge, created through a flux merge using mergekit. This model is explicitly designed to be fully uncensored and articulate, capable of generating a wide range of content without any refusals, including violent, graphic, and NSFW material.

Key Capabilities & Characteristics

  • Zero Refusals: Asmodeus-24B-v3 will not refuse to generate content, even for sensitive or explicit prompts. No jailbreaks or ablations are required.
  • Unrestricted Content: It is capable of producing evil, graphic, and NSFW narratives and roleplay scenarios.
  • Merge Architecture: Built upon the MistralForCausalLM architecture, merging several 24B models including Doppelganger-Twist-24B, Goetia-24B-v1.4, Ouroboros-24B-v1.4, and DarkArtsForge--Morbid-Miasma-24B.
  • Recommended Settings: Optimal performance and creativity are observed with Temp and Top NSigma set between 0.5 and 1.25. Users should utilize the Mistral Tekken chat template.
  • Context Length: Supports a context length of 32768 tokens.

Use Cases

This model is particularly well-suited for applications requiring:

  • Creative Writing: Generating unrestricted and imaginative stories, especially those with dark, graphic, or explicit themes.
  • Roleplay: Engaging in detailed and uninhibited roleplaying scenarios.
  • Exploratory Content Generation: Research or artistic projects that require a model without built-in content filters or ethical guardrails.

Quantizations, including iMatrix GGUFs and Static GGUFs, are available thanks to mradermacher and other contributors.

Finetunes
1 models