NarutoDolphin-10B: A Merged Language Model

NarutoDolphin-10B is a 10.7 billion parameter large language model developed by FelixChao. This model is a strategic merge of two distinct 7B parameter models: FelixChao/WizardDolphin-7B and FelixChao/NinjaDolphin-7B.

Key Characteristics

• Parameter Count: 10.7 billion parameters, offering a balance between performance and computational requirements.
• Context Length: Supports a context window of 4096 tokens, enabling processing of moderately long inputs and generating coherent responses.
• Origin: Created through a merge operation, combining the capabilities and knowledge bases of its two base models.

Deployment Options

• Quantized Versions: Optimized, quantized versions of NarutoDolphin-10B are available, thanks to contributions from s3nh. These versions, such as s3nh/NarutoDolphin-10B-GGUF, are designed for more efficient inference on various hardware.

Usage

This model can be readily integrated into applications using the Hugging Face transformers library, with provided Python code examples for text generation tasks.