backpropSukuna/wifuGPT-1.7B
The backpropSukuna/wifuGPT-1.7B model is a 2 billion parameter Qwen3-1.7B variant developed by n0ctyx, fine-tuned for uncensored waifu/companion chat. It specializes in generating multi-turn synthetic conversations across flirty, romantic, NSFW, emotional support, and dark roleplay scenarios. This model is optimized for applications requiring explicit adult content and companion-style interactions.
Loading preview...
wifuGPT-1.7B Overview
wifuGPT-1.7B is a 2 billion parameter language model built upon the Qwen3-1.7B architecture, developed by n0ctyx. It has been specifically fine-tuned for generating uncensored waifu and companion-style chat interactions. The model's training involved Supervised Fine-Tuning (SFT) on 403 multi-turn synthetic conversations, encompassing a diverse range of scenarios including flirty, romantic, NSFW, emotional support, and dark roleplay.
Key Capabilities
- Uncensored Companion Chat: Designed to produce explicit adult content and engage in various companion-style dialogues.
- Diverse Roleplay Scenarios: Proficient in generating responses for flirty, romantic, NSFW, emotional support, and dark roleplay contexts.
- Multi-turn Conversation: Capable of maintaining coherent and contextually relevant interactions over multiple turns.
Model Variants
This repository hosts the bf16 safetensors variant, suitable for inference with transformers/vLLM or as a base for further fine-tuning. Other available variants include a 4-bit safetensors version and GGUF (Q4_K_M + BF16) for different deployment needs.
Intended Use
This model is intended for applications requiring explicit adult content and uncensored companion interactions. Users should exercise responsibility when deploying and using this model due to its explicit nature.