cyan1143/Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus

TEXT GENERATIONPricing:Input $0.87 / Cached $0.2 / Output $0.99Concurrent Unit Cost:1Model Size:12BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 28, 2026Architecture:Transformer Featherless Exclusive Cold

The cyan1143/Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus is a 12 billion parameter instruction-tuned Mistral Nemo model, fine-tuned for enhanced reasoning and "thinking" capabilities, leveraging a Claude Opus 4.5 high-reasoning dataset. This model is uncensored and optimized for generating detailed, complex, and high-quality outputs across various tasks, including vivid prose and intense narratives. It features a compact reasoning engine that improves performance without being overly verbose, supporting a context length of up to 32768 tokens.

Loading preview...

Model Overview

The cyan1143/Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus is a 12 billion parameter instruction-tuned model based on Mistral Nemo. It has been specifically fine-tuned using an Unsloth process with a Claude Opus 4.5 high-reasoning dataset, transforming it into a "thinking/reasoning" model. A key differentiator is its "Heretic'ed" status, meaning it is significantly de-censored (refusal rate reduced from 87/100 to 14/100) prior to the reasoning fine-tuning, resulting in a fully uncensored and highly capable model.

Key Capabilities

  • Enhanced Reasoning: Integrates a compact reasoning/thinking engine, generating 3-6 paragraph (300-600 token) reasoning blocks that improve overall output quality, detail, length, and complexity.
  • Uncensored Output: Provides unfiltered and uncensored responses, suitable for vivid prose, intense narratives, and R-18 horror content, including swearing and visceral details.
  • Flexible Temperature Settings: Reasoning activation is not affected by temperature, allowing for a wide range of temp settings (0.1 to 2.5 or higher) to control output creativity.
  • Optimized for Detail: The reasoning process directly enhances the generation of detailed and complex content, making it suitable for intricate storytelling and analytical tasks.
  • Context Length: Supports a substantial context window of 32768 tokens, with a suggested minimum of 8k+ for optimal performance.

Use Cases

This model is particularly well-suited for applications requiring:

  • Creative Writing: Generating detailed, vivid, and uncensored narratives, including horror, romance, and complex storytelling.
  • Roleplay: Providing dynamic and unfiltered responses for immersive role-playing scenarios.
  • Complex Problem Solving: Leveraging its enhanced reasoning capabilities to tackle intricate prompts and generate well-thought-out solutions.
  • Unrestricted Content Generation: For users who require a model without built-in content filters or "nannies" for specific use cases.

Finetunes
1 models