UNIVA-Bllossom/DeepSeek-llama3.3-Bllossom-70B
TEXT GENERATIONConcurrency Cost:4Model Size:70BQuant:FP8Ctx Length:32kPublished:Feb 12, 2025License:mitArchitecture:Transformer0.1K Open Weights Cold

The UNIVA-Bllossom/DeepSeek-llama3.3-Bllossom-70B is a 70 billion parameter language model developed by UNIVA and Bllossom, based on DeepSeek-R1-distill-Llama-70B. It is specifically post-trained to enhance reasoning performance in Korean environments, addressing limitations of its base model which was primarily trained on English and Chinese data. This model improves Korean inference by performing internal reasoning in English and generating responses in the input language, making it suitable for complex reasoning tasks in Korean.

Loading preview...