InfiX-ai/InfiR-1B-Base

InfiR-1B-Base is a 1 billion parameter base language model developed by InfiX, continually pretrained from Meta's Llama-3.2-1B. It features an extended 4k context length and is specifically optimized for reasoning tasks, including mathematical problem-solving and code generation. The model aims to improve reasoning capabilities in smaller models, making advanced AI more accessible and addressing privacy concerns.