yswill/llama-13b-hf
The yswill/llama-13b-hf model is a 13 billion parameter LLaMA model, provided in Hugging Face format for direct use with the Hugging Face API. This model is suitable for applications requiring a robust base LLM, including serving as the foundational LLaMA model for projects like LLaVA. Access to this model is granted under a non-commercial license, requiring prior approval.
Loading preview...
Model Overview
The yswill/llama-13b-hf model provides the weights for the 13 billion parameter LLaMA model, specifically formatted for direct integration and use with the Hugging Face API. This makes it readily accessible for developers who have been granted access to the original LLaMA weights but require them in a convenient, pre-converted format.
Key Characteristics
- Model Family: LLaMA
- Parameter Count: 13 billion parameters
- Format: Hugging Face (HF) compatible, allowing for straightforward loading and utilization.
- License: Non-commercial, requiring prior access approval from the original LLaMA distribution.
Use Cases
This model is particularly well-suited for:
- Foundational LLM: Serving as a powerful base language model for various natural language processing tasks.
- LLaVA Integration: It is explicitly mentioned as being suitable for use as the underlying LLaMA model in multimodal projects such as LLaVA.
- Research and Development: For researchers and developers with approved access to LLaMA, this provides a convenient way to work with the 13B variant in a standard framework.