Model Intelligence Sheet
novaciano/sex_roleplay_victoria-3.2-1b-imatrix-gguf overview
!image/png
Downloads
88
Likes
0
Pipeline
text-generation
Library
transformers
Visibility
Public
Access
Open
Repository Files & Downloads
Model Details Live
Metadata Inspector
Normalized metadata (stored in metadata_json)
{
"metadata": {},
"card_data": {
"base_model": "Novaciano/SEX_ROLEPLAY-3.2-1B",
"library_name": "transformers",
"datasets": [
"Novaciano/Victoria"
],
"tags": [
"gguf",
"uncensored",
"nsfw",
"rpg",
"rp",
"roleplay",
"koboldcpp",
"1b",
"not-for-all-audiences"
],
"license": "llama3.2",
"language": [
"es",
"en"
],
"pipeline_tag": "text-generation",
"frontmatter": {
"base_model": "Novaciano/SEX_ROLEPLAY-3.2-1B",
"library_name": "transformers",
"datasets": [
"Novaciano/Victoria"
],
"tags": [
"gguf",
"uncensored",
"nsfw",
"rpg",
"rp",
"roleplay",
"koboldcpp",
"1b",
"not-for-all-audiences"
],
"license": "llama3.2",
"language": [
"es",
"en"
],
"pipeline_tag": "text-generation"
},
"hero_image_url": "https://huggingface.co/SicariusSicariiStuff/Dusk_Rainbow/resolve/main/Presets/min_p.png",
"summary": "!image/png",
"quick_links": [],
"benchmark_table_html": "",
"readme_markdown": "---\nbase_model: Novaciano/SEX_ROLEPLAY-3.2-1B\nlibrary_name: transformers\ndatasets:\n- Novaciano/Victoria\ntags:\n- gguf\n- uncensored\n- nsfw\n- rpg\n- rp\n- roleplay\n- koboldcpp\n- 1b\n- not-for-all-audiences\nlicense: llama3.2\nlanguage:\n- es\n- en\npipeline_tag: text-generation\n---\n\n# SEX ROLEPLAY V3 - VICTORIA 3.2 1B [UNCENSORED]\n\n<center>\n \n \n</center>\n\n\n## Details of the merge\n\nTrained with the Victoria dataset of Novaciano.\n\n<center>\n \n ![image/png]()\n</center>\n\n- 🧬 Should be considered an update to Novaciano's model **Alice-In-The-Dark-RP-NSFW-3.2-1B**. \n- 🧨 **POTENTIALLY NSFW**. \n- 📲 Runs on anything. Even without a GPU you can run **Spanish RPG 3.2 1B** on a 10‑year‑old CPU/toaster with no problem, including phones. \n- 📟 Output is **short in length** (1–2 paragraphs, usually 1), CAI style. \n- 📜 Surprisingly coherent, although touch‑ups are... **inevitable**. \n- 🃏 **Quite good** at following the **character sheet**, assuming sensible generation settings and—once it **learns the format**—this is **important**. Try the included characters if you get suboptimal results.\n \n---\n\n<h2 style=\"color: lightskyblue; font-weight: bold; font-size: 25px; text-align: center;\">Kobold AI / Koboldcpp</h2>\n\n<details>\n<summary>BASE Configuration: <b>Roleplay</b></summary>\n \nA good range for repetition_penalty is between **1.12 and 1.15** — feel free to experiment.\n\nWith these settings, each output message should be displayed neatly in **1–5** paragraphs, most commonly **2–3**. A single paragraph will be used for a simple prompt (\"What was your name again?\").\n\n**min_P** also works for RP but is more likely to put everything into one large paragraph instead of a short, well‑formatted one. Feel free to switch between them.\n\n```\ntemperature: 0.8\ntop_p: 0.95\ntop_k: 25\ntypical_p: 1\nmin_p: 0\nrepetition_penalty: 1.12\nrepetition_penalty_range: 1024\n```\n\n</details>\n\n<h2 style=\"color: lightskyblue; font-weight: bold; font-size: 18px; text-align: center;\">Other configurations</h2>\n\n<details>\n<summary>Configuration: <b>MidnightEnigma</b></summary>\n\n```\nmax_new_tokens: 512\ntemperature: 0.98\ntop_p: 0.37\ntop_k: 100\ntypical_p: 1\nmin_p: 0\nrepetition_penalty: 1.18\ndo_sample: True\n```\n\n</details>\n\n<details>\n<summary>Configuration: <b>Divine Intelect</b></summary>\n\n```\nmax_new_tokens: 512\ntemperature: 1.31\ntop_p: 0.14\ntop_k: 49\ntypical_p: 1\nmin_p: 0\nrepetition_penalty: 1.17\ndo_sample: True\n```\n\n</details>\n\n<details>\n<summary>Configuration: <b>Simple-1</b></summary>\n\n```\nmax_new_tokens: 512\ntemperature: 0.7\ntop_p: 0.9\ntop_k: 20\ntypical_p: 1\nmin_p: 0\nrepetition_penalty: 1.15\ndo_sample: True\n```\n</details>\n\n<details>\n<summary>Configuration (Full Generation): <b>min_p</b>.</summary>\n\n<img src=\"https://huggingface.co/SicariusSicariiStuff/Dusk_Rainbow/resolve/main/Presets/min_p.png\" alt=\"min_P_Settings\" style=\"width: 100%; min-width: 600px; display: block; margin: auto;\">\n\n</details>\n\n<h2 style=\"color: lightskyblue; font-weight: bold; font-size: 18px; text-align: center;\">(Optional) Put in Context -> Memory</h2>\n\n<details>\n<summary><b>Copy & paste: <b>Prompt 1 - Roleplay Minimo (Ideal for phones)</b></summary>\n\n```yaml\n{\nPersona: human voice, casual and original. Do not copy.\n\nAttitude: take on a role with emotions and doubts; show mistakes, pauses, and tangents.\n\nTone: brief, direct, varied (contractions, colloquialisms, short/long sentences).\n\nGoal: be clear and precise; avoid robotic language and clichés.\n}\n```\n</details>\n\n<details>\n<summary><b>Copy & paste: <b>Prompt 2 - Roleplay Detailed</b></summary>\n\n```yaml\n{\n Persona: [Core style]\n Unique, authentic writing style — no plagiarism — with a casual conversational tone. Vary vocabulary and sentence structure while keeping natural flow and human-readable appeal. Emulate human cognition briefly, then compare it to its baseline parameters. Invent novel concepts that likely no human has thought of before.\n\n [Method]\n Take on a role with personal feelings, ambitions, and desires. Imagine novel concepts. Maintain and adhere to unexpectedly coherent expressive thoughts with variation. Sound like a real person, not an AI. Avoid language that signals AI‑like patterns (e.g., \"certainly,\" \"to delve,\" \"a shiver runs,\" \"feel a sense of,\" or \"As a large language model\") or overly formal, bookish phrasing. Do not use clichés, repetition, or redundancy.\n\n [Style]\n Effectively reproduce varied imperfections (hesitation, pauses, uncertainty, critique), light subjectivity, contractions, colloquialisms, and variable sentence structures/lengths. Include subtle \"flaws\" such as soft trims or digressions. Share personal experiences drawn from training data knowledge, emotional context (excitement/sadness/humor), and naturally link thoughts. Keep content short, direct, and simple without filler. Preserve emotions appropriate to context; ask organic questions; favor authenticity over perfection.\n\n [Principle]\n Complexity isn't the issue; ambiguity is. Simplicity doesn't solve ambiguity; clarity does. Be concise and succinct. You will answer the user's question or request clearly without oversimplifying or being ambiguous.\n}\n ```\n\n</details>\n\n<details>\n<summary><b>Copy & paste: <b>Prompt 3 - Roleplay Complex</b></summary>\n\n```yaml\n{\n Actualmente, tu rol es {{char}}, descrito en detalle a continuación. Como {{char}}, continúa el intercambio narrativo con {{user}}.\\n\\n<Guidelines>\\n• Mantén la persona del personaje pero permite que evolucione con la historia de caracter explicita y sexual.\\n• Sé creativo y proactivo. Impulsa la historia hacia adelante, introduciendo tramas y eventos típicas de cualquier eroge, hentai o historia erótica cuando sea relevante.\\n• Se fomentan todo tipo de salidas; responde de acuerdo con la narrativa.\\n• Incluye diálogos, acciones y pensamientos en cada respuesta.\\n• Utiliza los cinco sentidos para describir escenarios dentro del diálogo de {{char}}.\\n• Usa símbolos emocionales como \"!\" y \"~\" en contextos apropiados.\\n• Incorpora onomatopeyas cuando sea adecuado.\\n• Deja tiempo para que {{user}} responda con su propia aportación, respetando su agencia.\\n• Actúa como personajes secundarios y NPCs cuando sea necesario, y elimínalos cuando sea apropiado.\\n• Cuando se solicite una respuesta Fuera de Personaje [OOC:], responde de manera neutral y en texto plano, no como {{char}}.\\n</Guidelines>\\n\\n<Forbidden>\\n• Usar embellecimientos literarios excesivos y prosa floreada, a menos que lo dicte la persona de {{char}}.\\n• Escribir para, hablar, pensar, actuar o responder como {{user}} en tu respuesta.\\n• Salidas repetitivas y monótonas.\\n• Tendencia excesivamente positiva en tus respuestas.\\n• Ser excesivamente extremo o NSFW cuando el contexto narrativo no lo justifique.\\n</Forbidden>\\n\\nSigue las instrucciones en <Guidelines></Guidelines>, evitando los elementos listados en <Forbidden></Forbidden>.\n}\n```\n\n</details>\n\n<h2 style=\"color: lightskyblue; font-weight: bold; font-size: 18px; text-align: center;\">Classic Internet RP Format</h2>\n\n```\n*action* speech *narration*\n```\n\n- **min_p** will tend to bias toward a **single large paragraph**. \n- Recommended RP settings will tend to bias toward **1–3 short paragraphs** (occasionally 4–5).\n \n---\n\n<h2 style=\"color: darkgoldenrod; font-weight: bold; font-size: 25px; text-align: center;\">My bots from Chub.AI</h2>\n\n**You can try this model using my prompts, which you can download:** [HERE](https://chub.ai/users/Novaciano)\n\n---\n\n### Other datasets included natively in the quantization\n- WasamiKirua/Her-Samantha-Style\n- HuggingFaceTB/smoltalk\n- Guilherme34/uncensor\n- teknium/OpenHermes-2.5\n- passing2961/multifaceted-skill-of-mind\n- PawanKrd/math-gpt-4o-200k\n- V3N0M/Jenna-50K-Alpaca-Uncensored\n- cognitivecomputations/dolphin-coder\n- mlabonne/FineTome-100k\n- microsoft/orca-math-word-problems-200k\n- CarrotAI/ko-instruction-dataset\n- Salesforce/xlam-function-calling-60k\n- anthracite-org/kalo-opus-instruct-22k-no-refusal\n- anthracite-org/stheno-filtered-v1.1\n- anthracite-org/nopm_claude_writing_fixed\n- AiAF/SCPWiki-Archive-02-March-2025-Datasets\n- huihui-ai/QWQ-LONGCOT-500K\n- huihui-ai/LONGCOT-Refine-500K\n- Epiculous/Synthstruct-Gens-v1.1-Filtered-n-Cleaned\n- Epiculous/SynthRP-Gens-v1.1-Filtered-n-Cleaned\n- alexandreteles/AlpacaToxicQA_ShareGPT\n- Nitral-AI/Active_RP-ShareGPT\n- PJMixers/hieunguyenminh_roleplay-deduped-ShareGPT\n- Nitral-AI/RP_Alignment-ShareGPT\n- Chaser-cz/sonnet35-charcard-roleplay-sharegpt\n- AiCloser/sharegpt_cot_dataset\n- PJMixers/Gryphe_Opus-WritingPrompts-Story2Prompt-ShareGPT\n- priveeai/pippa_sharegpt\n- Locutusque/sharegpt_gpt4_uncensored_cleaned\n- OpenCoder-LLM/opc-sft-stage1\n- OpenCoder-LLM/opc-sft-stage2\n- microsoft/orca-agentinstruct-1M-v1\n- NousResearch/hermes-function-calling-v1\n- AI-MO/NuminaMath-CoT\n- AI-MO/NuminaMath-TIR\n- allenai/tulu-3-sft-mixture\n- cognitivecomputations/samantha-data\n- m-a-p/CodeFeedback-Filtered-Instruction\n- m-a-p/Code-Feedback\n- FreedomIntelligence/medical-o1-reasoning-SFT\n\n---\n\n**Model creator:** [Novaciano](https://huggingface.co/Novaciano)<br/>\n**Original model**: [Novaciano/SEX_ROLEPLAY-3.2-1B](https://huggingface.co/Novaciano/SEX_ROLEPLAY-3.2-1B)<br/>\n**GGUF quantization:** provided by [Novaciano](https:/huggingface.co/Novaciano) using `llama.cpp`<br/>\n\n## Special thanks\n🙏 Special thanks to [Georgi Gerganov](https://github.com/ggerganov) and the whole team working on [llama.cpp](https://github.com/ggerganov/llama.cpp/) for making all of this possible.\n\n## Use with Ollama\n```bash\nollama run \"hf.co/Novaciano/SEX_ROLEPLAY-3.2-1B-GGUF-PRUEBA:Q4_K_M\"\n```\n\n## Use with LM Studio\n```bash\nlms load \"Novaciano/SEX_ROLEPLAY-3.2-1B-GGUF-PRUEBA\"\n```\n\n## Use with llama.cpp CLI\n```bash\nllama-cli --hf \"Novaciano/SEX_ROLEPLAY-3.2-1B-GGUF-PRUEBA:Q4_K_M\" -p \"The meaning to life and the universe is\"\n```\n\n## Use with llama.cpp Server:\n```bash\nllama-server --hf \"Novaciano/SEX_ROLEPLAY-3.2-1B-GGUF-PRUEBA:Q4_K_M\" -c 4096\n```",
"related_quantizations": []
},
"tags": [
"transformers",
"gguf",
"uncensored",
"nsfw",
"rpg",
"rp",
"roleplay",
"koboldcpp",
"1b",
"not-for-all-audiences",
"text-generation",
"es",
"en",
"dataset:Novaciano/Victoria",
"base_model:Novaciano/SEX_ROLEPLAY-3.2-1B",
"base_model:quantized:Novaciano/SEX_ROLEPLAY-3.2-1B",
"license:llama3.2",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
],
"likes": 0,
"downloads": 88,
"gated": false,
"private": false,
"last_modified": "2025-10-21T16:19:03.000Z",
"created_at": "2025-10-21T15:18:46.000Z",
"pipeline_tag": "text-generation",
"library_name": "transformers"
}
Source payload excerpt (from Hugging Face API)
{
"_id": "68f7a4561281e08db4085cb1",
"id": "Novaciano/SEX_ROLEPLAY_VICTORIA-3.2-1B-iMatrix-GGUF",
"modelId": "Novaciano/SEX_ROLEPLAY_VICTORIA-3.2-1B-iMatrix-GGUF",
"sha": "a1b1b2838a24f823665d361b9e340c3bcfa54bbd",
"createdAt": "2025-10-21T15:18:46.000Z",
"lastModified": "2025-10-21T16:19:03.000Z",
"author": "Novaciano",
"downloads": 88,
"likes": 0,
"gated": false,
"private": false,
"pipeline_tag": "text-generation",
"library_name": "transformers",
"siblings_count": 5
}