SakanaAI/TinySwallow-1.5B-Instruct

TinySwallow-1.5B-Instruct is a 1.5 billion parameter instruction-tuned causal language model developed by Sakana AI, specifically optimized for Japanese language tasks. It was created using Temporally Adaptive Interpolated Distillation (TAID) with Qwen2.5-32B-Instruct as the teacher model and Qwen2.5-1.5B-Instruct as the student. This model excels at following instructions and engaging in conversations in Japanese, leveraging its 131072 token context length for complex interactions.