avans06/Meta-Llama-3.2-8B-Instruct

avans06/Meta-Llama-3.2-8B-Instruct is an 8 billion parameter text-only language model derived from Meta's Llama-3.2-11B-Vision-Instruct. This model was created by removing the vision layer and cross-attention layers from the original multimodal architecture, resulting in a more compact model optimized exclusively for text-based tasks. It retains the Llama 3.2 architecture's strengths for general language understanding and generation, making it suitable for applications requiring efficient text processing without visual input.