rinna/qwq-bakeneko-32b

rinna/qwq-bakeneko-32b is a 32.8 billion parameter instruction-tuned reasoning model developed by rinna, based on the Qwen2.5 architecture. It is fine-tuned using Chat Vector and Odds Ratio Preference Optimization (ORPO) to deliver superior performance in Japanese language tasks. This model is specifically designed for reasoning applications and adheres to the Qwen/QwQ-32B chat format. It offers a 131072 token context length, making it suitable for complex Japanese language processing.