DandelionConvention/Qwen3-Sex

TEXT GENERATIONConcurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 11, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

DandelionConvention/Qwen3-Sex is a 4 billion parameter instruction-tuned dialogue model based on Qwen/Qwen3-4B-Instruct-2507. Fine-tuned for adult-oriented content, it generates explicit and sensitive text. This bfloat16 model supports a native context length of 262,144 tokens, though 4K-32K is recommended for general dialogue. It is specifically designed for academic research, non-commercial personal use, and fictional creation involving mature themes.

Loading preview...

Overview

DandelionConvention/Qwen3-Sex is a 4 billion parameter dialogue model derived from the Qwen/Qwen3-4B-Instruct-2507 base model. It has undergone Supervised Fine-Tuning (SFT) with a focus on generating adult-oriented content, including explicit descriptions and sensitive text. The model operates in bfloat16 precision and natively supports a substantial context length of 262,144 tokens, though a range of 4K to 32K tokens is suggested for typical conversational use cases.

Key Capabilities

  • Adult Content Generation: Specifically fine-tuned to produce text containing adult themes and explicit descriptions.
  • Dialogue Optimization: Designed for multi-turn conversational interactions, leveraging the ChatML template (<|im_start|> / <|im_end|>).
  • Flexible Deployment: Provided with safetensors weights, GGUF formats (BF16 and Q8_0 quantized), and an Ollama Modelfile for various deployment environments.
  • Optimized Inference: Includes a generation_config.json with recommended parameters (e.g., temperature: 0.7, top_p: 0.8, repetition_penalty: 1.08) for optimal output quality.

Intended Use Cases

This model is intended for users aged 18 and above and is suitable for:

  • Academic Research: Studying the generation of adult-oriented content in LLMs.
  • Personal Non-Commercial Use: For individual exploration and experimentation.
  • Fictional Creation: Assisting in the generation of creative writing that includes mature themes.

Users are strongly cautioned that the model may produce content unsuitable for all audiences and must comply with all local laws and regulations. Quality may vary between safetensors and GGUF formats, with safetensors generally offering superior detail and coherence.

Finetunes
1 models