zoom_in Click to enlarge

Yi-34B

▲ ▼

language

description Yi-34B Overview

Yi-34B is a 34-billion-parameter language model developed by 01.AI, an artificial intelligence company founded by Kai-Fu Lee, and released in 2023. The model is explicitly designed as bilingual, with strong capabilities in both English and Chinese across reasoning, mathematics, coding, and general knowledge tasks. Yi-34B was released with open weights to encourage research and development, demonstrating competitive performance against other models in its parameter range. The model represents 01.AI's commitment to advancing foundation models specifically optimized for the Chinese and global markets.

help Yi-34B FAQ

Who created the Yi-34B language model?

Yi-34B was developed by 01.AI, a Chinese AI startup founded by Kai-Fu Lee, who previously led Google China and Microsoft Research Asia. The model was released in 2023 as part of 01.AI's Yi series of foundation models.

Is Yi-34B good at Chinese language tasks?

Yi-34B was specifically designed as a bilingual model optimized for both English and Chinese, and it demonstrated competitive performance against models like Llama 2 on Chinese-language benchmarks. Its training data included substantial Chinese corpora, giving it an advantage over models primarily trained on English text.

Can I download and use Yi-34B model weights?

01.AI released Yi-34B weights openly on Hugging Face, though the license includes certain restrictions that differ from fully permissive licenses like Apache 2.0. Users should review the specific terms on the model's Hugging Face page before commercial use.

Are there smaller Yi models besides the 34B version?

Yes, 01.AI released multiple sizes in the Yi family, including a 6-billion-parameter variant (Yi-6B) alongside the larger Yi-34B. The smaller models are intended for researchers and developers who need lower computational requirements while retaining the bilingual English and Chinese capabilities.

swap_horiz

Looking for Yi-34B alternatives? Compare top competitors ranked & scored

arrow_forward