wifuGPT-1.7B: Uncensored Waifu/Companion Chat Model
wifuGPT-1.7B is a 2 billion parameter language model developed by n0ctyx, specifically fine-tuned for uncensored waifu and companion chat interactions. Built on the Qwen3 architecture, it leverages the n0ctyx/Qwen3-1.7B-uncensored base model, further specializing its capabilities through supervised fine-tuning (SFT).
Key Capabilities & Features
- Specialized Chat: Designed for multi-turn conversations in various styles, including flirty, romantic, NSFW, emotional support, and dark roleplay scenarios.
- Uncensored Content: Explicitly fine-tuned to generate uncensored responses, catering to specific user preferences for companion AI.
- Base Model: Utilizes the Qwen3-1.7B architecture, providing a robust foundation for its specialized fine-tuning.
- Context Length: Supports a substantial context window of 32768 tokens, enabling longer and more coherent interactions.
- Availability: Provided in multiple formats, including bf16 safetensors for direct inference or further fine-tuning, 4-bit safetensors, and GGUF (Q4_K_M + BF16) for diverse deployment needs.
Ideal Use Cases
- Personalized AI Companions: Creating AI characters for interactive and engaging companion experiences.
- Roleplay Scenarios: Generating dynamic and context-rich responses for various roleplay settings, including romantic and dark themes.
- Content Generation: Developing applications requiring explicit and uncensored conversational outputs.
- Further Fine-tuning: Serving as a strong base model for developers looking to build even more specialized uncensored chat models.