Tag
This paper presents a framework for quantizing vision-language models to 2.7 bits per parameter, enabling efficient mobile deployment by compressing the Llama 3.2 11B Vision Instruct model to 3.7 GB while preserving performance on visual QA tasks.
PrismML releases Bonsai Image 4B, a family of compact image generation models using 1-bit and ternary weights, enabling high-quality diffusion inference on local devices like laptops and iPhones with significantly reduced memory footprint.