wkplhc/qwen3.5-4b-nsfw-ara-heretic-literotica-i1-gguf overview

{
  "metadata": {},
  "card_data": {
    "language": [
      "en"
    ],
    "base_model": "Sinbad-The-Sailor/Qwen3.5-4B-NSFW-ARA-Heretic-Literotica",
    "pipeline_tag": "text-generation",
    "tags": [
      "nsfw",
      "story-generation",
      "erotica",
      "literotica",
      "qwen",
      "lora",
      "ara",
      "heretic",
      "uncensored",
      "gguf",
      "i1",
      "imatrix"
    ],
    "frontmatter": {
      "language": [
        "en"
      ],
      "base_model": "Sinbad-The-Sailor/Qwen3.5-4B-NSFW-ARA-Heretic-Literotica",
      "pipeline_tag": "text-generation",
      "tags": [
        "nsfw",
        "story-generation",
        "erotica",
        "literotica",
        "qwen",
        "lora",
        "ara",
        "heretic",
        "uncensored",
        "gguf",
        "i1",
        "imatrix"
      ]
    },
    "hero_image_url": "",
    "summary": "**The i1 release introduces models quantized with an Importance Matrix, significantly improving performance on key prompt structures.**",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlanguage:\n- en\nbase_model: Sinbad-The-Sailor/Qwen3.5-4B-NSFW-ARA-Heretic-Literotica\npipeline_tag: text-generation\ntags:\n- nsfw\n- story-generation\n- erotica\n- literotica\n- qwen\n- lora\n- ara\n- heretic\n- uncensored\n- gguf\n- i1\n- imatrix\n---\n\n# Qwen3.5-4B-NSFW-ARA-Heretic-Literotica-i1-GGUF\n\n**The `i1` release introduces models quantized with an Importance Matrix, significantly improving performance on key prompt structures.**\n\n## Overview\nThis repository contains GGUF (GPT-Generated Unified Format) versions of the **[Sinbad-The-Sailor/Qwen3.5-4B-NSFW-ARA-Heretic-Literotica](https://huggingface.co/Sinbad-The-Sailor/Qwen3.5-4B-NSFW-ARA-Heretic-Literotica)** model, a specialized language model built on the **Qwen 3.5 architecture (4B parameters)**. It is designed for immersive erotic storytelling and creative prose, inheriting the technical uncensorship approaches of the **ARA** and **Heretic** frameworks.\n\nThese GGUF models have been quantized using an **Importance Matrix (Imatrix)**, making them more robust and preserving key knowledge that is often lost in standard quantization.\n\n### What is GGUF?\nGGUF is a binary format designed for single-file deployment of large language models, making it easy to use with tools like `llama.cpp`. It is a successor to the GGML format and offers better performance, flexibility, and metadata support.\n\n### The `i1` Imatrix Quantization\nThe `.i1.` in the filenames signifies that these models were quantized using an **Importance Matrix**. This advanced technique measures the sensitivity of different weights in the neural network and uses this information to quantize the model more intelligently.\n\n**Benefits:**\n- **Reduced Perplexity**: Imatrix quantization significantly lowers perplexity compared to standard methods.\n- **Preserved Knowledge**: It helps retain critical information and nuances that are often damaged during the quantization process.\n- **Higher Quality**: The resulting models provide more coherent and contextually accurate outputs, especially for complex creative tasks.\n\n## Available Models\nThe following quantized models are available, offering a trade-off between performance and resource usage:\n\n| File Name                                                                    | Quantization      | Recommended Use                               |\n| ---------------------------------------------------------------------------- | ----------------- | --------------------------------------------- |\n| `Qwen3.5-4B-NSFW-ARA-Heretic-Literotica.i1.Q4_K_M.gguf`                        | **Q4_K_M**        | Balanced quality, good for most use cases.    |\n| `Qwen3.5-4B-NSFW-ARA-Heretic-Literotica.i1.Q5_K_M.gguf`                        | **Q5_K_M**        | High-quality, recommended for creative prose. |\n| `Qwen3.5-4B-NSFW-ARA-Heretic-Literotica.i1.Q6_K.gguf`                        | **Q6_K**          | Very high quality, for demanding tasks.       |\n\n*These files were renamed to include the `.i1.` tag to signify the Imatrix quantization. You may need to rename them locally if you have already downloaded the old files.*\n\n## Usage & Inference\nThese models are designed to be used with `llama.cpp` and other GGUF-compatible backends.\n\n### Example with `llama.cpp`:\n1.  **Download `llama.cpp`**:\n    ```bash\n    git clone https://github.com/ggerganov/llama.cpp.git\n    cd llama.cpp\n    make\n    ```\n2.  **Download a GGUF model** from this repository.\n3.  **Run inference**:\n    ```bash\n    ./main -m <path_to_gguf_model> -p \"USER: [Your prompt here]\nASSISTANT:\" -n 256 --temp 0.8 -c 4096\n    ```\n\n### Prompt Format\nThis model uses the standard **ChatML** format, adapted for a raw GGUF interface:\n<|im_start|>system\nYou are a creative and descriptive erotic storyteller.<|im_end|>\n<|im_start|>user\n[Your prompt here]<|im_end|>\n<|im_start|>assistant\n\nWhen using a text-based interface like `llama.cpp`, you can format your prompt like this:\n\n```\n<|im_start|>system\nYou are a creative and descriptive erotic storyteller.<|im_end|>\n<|im_start|>user\nShe walked into the room, her eyes sparkling with a mischievous glint.<|im_end|>\n<|im_start|>assistant\n```\n\n### Inference Settings\nFor optimal creative output, the following parameters are recommended:\n- **Temperature**: 0.75 - 0.85\n- **Top-p**: 0.90\n- **Repetition Penalty**: 1.1\n\n## Responsible Use & Disclaimer\nThis model is intended for adult audiences (18+) and generates explicit NSFW content. It is built on uncensored foundations (ARA/Heretic) and is designed to follow creative prompts without artificial refusal. Users are responsible for ensuring their use of the model complies with local laws and regulations. The model is a tool for creative fiction and should not be used to generate illegal or non-consensual content.\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "nsfw",
    "story-generation",
    "erotica",
    "literotica",
    "qwen",
    "lora",
    "ara",
    "heretic",
    "uncensored",
    "i1",
    "imatrix",
    "text-generation",
    "en",
    "base_model:Sinbad-The-Sailor/Qwen3.5-4B-NSFW-ARA-Heretic-Literotica",
    "base_model:adapter:Sinbad-The-Sailor/Qwen3.5-4B-NSFW-ARA-Heretic-Literotica",
    "endpoints_compatible",
    "region:us",
    "conversational",
    "not-for-all-audiences"
  ],
  "likes": 0,
  "downloads": 889,
  "gated": false,
  "private": false,
  "last_modified": "2026-04-01T00:17:03.000Z",
  "created_at": "2026-04-01T00:10:20.000Z",
  "pipeline_tag": "text-generation",
  "library_name": ""
}

{
  "_id": "69cc626ca4846554fb78bf01",
  "id": "wkplhc/Qwen3.5-4B-NSFW-ARA-Heretic-Literotica-i1-GGUF",
  "modelId": "wkplhc/Qwen3.5-4B-NSFW-ARA-Heretic-Literotica-i1-GGUF",
  "sha": "c86711d557c1852e442da6648f99560ae12ed18d",
  "createdAt": "2026-04-01T00:10:20.000Z",
  "lastModified": "2026-04-01T00:17:03.000Z",
  "author": "wkplhc",
  "downloads": 889,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "text-generation",
  "library_name": "",
  "siblings_count": 3
}