Skip to content

EthenEthenEthen

Open Source Model Profile · DavidAU

MN-CaptainErisNebula-Chimera-v1.1-THINKING-ClaudeOpus4.5-12B-heretic-uncensored

MN-CaptainErisNebula-Chimera-v1.1-THINKING-ClaudeOpus4.5-12B-heretic-uncensored is a 12.25B-parameter Mistral text-generation fine-tune from DavidAU. Its card documents decensored reasoning tuning.

Publisher
DavidAU
Task
text-generation
Model type
mistral
License
Unknown
Library
transformers
Publication status
Accepted · not indexed

Model overview

This DavidAU release is published as a text-generation model. The captured configuration identifies MistralForCausalLM, and Safetensors metadata reports 12,247,782,400 parameters. According to the model card, it is a Heretic-decensored CaptainErisNebula Chimera variant trained for Claude Opus 4.5 style thinking and reasoning.

Recorded capabilities

Reasoning-oriented Chimera variant

According to the model card, the release adds Claude Opus 4.5 thinking and reasoning to the Heretic decensored Chimera line on Mistral Nemo 12B architecture.

Publisher-described uncensored behavior

According to the model card, the Heretic version is described as unlocked and unfiltered, with reported refusal and KL measurements.

Chat and sampler guidance

According to the model card, the publisher documents no-system-prompt use, Smoothing_factor 1.5 guidance for KoboldCpp and text-generation-webui, and a Class 1 settings guide.

Use cases in the source record

  • Creative-fiction, storytelling, and role-play text-generation workflows of the kind tagged and described in the model card.
  • Chat and role-play setups that follow the card's no-system-prompt note and sampler-smoothing guidance.

Limitations and unknowns

  • No license value was extracted from this record.
  • No evaluation results beyond the publisher's refusal and KL figures were extracted.
  • Context figures are publisher claims and were not independently verified.
  • Provider state is historical snapshot data, not independently refreshed current availability.
  • Uncensored-behavior and reasoning claims come from the publisher model card and were not independently verified.

Source and provenance

Source: DavidAU/MN-CaptainErisNebula-Chimera-v1.1-THINKING-ClaudeOpus4.5-12B-heretic-uncensored

Captured: Unknown. Processed: 2026-09-07T19:35:39.046830+00:00.

WARNING "HERETIC" version: Unlocked. UNFILTERED. NSFW. Vivid prose. INTENSE. Visceral Details. Light to R-18 HORROR. Swearing. UNCENSORED... humor, romance, fun... and UNFILTERED TRUTH. MN-CaptainErisNebula-Chimera-v1.1-THINKING-ClaudeOpus4.5-12B-heretic-uncensored I took the "Heretic" version (decensored) and trained it to add "Claude Opus 4.5" thinking and reasoning. (trained via Unsloth) This is a full reasoning/thinking, and fully uncensored model on Mistral Nemo 12B arch (max context 1 million, but suggest 128k to 256k max). This addition of "reasoning" signficantly elevates the power of this model. (3 example generations at th…

F001F002F003F004F005F006F008F009F010F011F012F013F014F015F018F019