SkylarWhite/SahabatAI-MediChatIndo-8B-v1-gguf

This is hosts various GGUF quantized versions of the gmonsoon/SahabatAI-MediChatIndo-8B-v1 model. This model is designed for medical and general-purpose conversational AI in Indonesian and is based on the LLaMA3 architecture. The GGUF format is optimized for efficient inference on low-resource devices and fast deployment.

Model Overview

SahabatAI-MediChatIndo-8B-v1 is a fine-tuned model created by merging:

• sethuiyer/Medichat-Llama3-8B
• GoToCompany/llama3-8b-cpt-sahabatai-v1-instruct

It has been optimized for understanding and responding in medical and general Indonesian conversations.

GGUF Quantized Versions

The following GGUF quantized versions are available in this repository:

• 16-bit (F16): High-precision quantization for use cases requiring maximal accuracy.
• Q4_K_M: Balanced between speed and performance, ideal for most use cases.
• Q5_K_M: Improved precision over Q4 while maintaining efficient performance.
• Q8_0: Full precision for demanding tasks where accuracy is critical.

Feedback and Contributions

Feedback and contributions are welcome! Please open an issue or contact the model's author for further discussions.