hsefz-ChenJunJie/Deepseek-R1-Distill-NSFW-RPv1-mlx-fp16
The hsefz-ChenJunJie/Deepseek-R1-Distill-NSFW-RPv1-mlx-fp16 is an 8 billion parameter language model, converted to MLX format by hsefz-ChenJunJie from the Deepseek-R1-Distill-NSFW-RPv1 base model. This model is specifically designed for NSFW (Not Safe For Work) and roleplay applications, leveraging its distilled architecture for focused performance. It is optimized for use within the Apple MLX ecosystem, providing efficient inference on Apple silicon.
Loading preview...
Model Overview
The hsefz-ChenJunJie/Deepseek-R1-Distill-NSFW-RPv1-mlx-fp16 is an 8 billion parameter language model, derived from the mergekit-community/Deepseek-R1-Distill-NSFW-RPv1 base model. This version has been specifically converted to the MLX format using mlx-lm version 0.22.3, making it suitable for efficient deployment and inference on Apple silicon.
Key Characteristics
- MLX Optimized: Converted for native execution on Apple's MLX framework, ensuring optimized performance on compatible hardware.
- Distilled Architecture: Based on a distilled version of the Deepseek-R1 model, suggesting a focus on efficiency while retaining key capabilities.
- Specialized for NSFW & Roleplay: The model's naming indicates a fine-tuning or design intent for applications involving Not Safe For Work content and roleplaying scenarios.
Usage
This model is intended for developers working within the MLX ecosystem. It can be loaded and used with the mlx-lm library for text generation tasks, including those requiring specific chat templates. The provided Python code snippet demonstrates how to load the model and tokenizer, apply chat templates if available, and generate responses.