by DeepSeek
DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is built upon the original V3 base checkpoint through a two-phase long context extension approach, following the methodology outlined in the original DeepSeek-V3 report. We have expanded our dataset by collecting additional long documents and substantially extending both training phases. The 32K extension phase has been increased 10-fold to 630B tokens, while the 128K extension phase has been extended by 3.3x to 209B tokens. Additionally, DeepSeek-V3.1 is trained using the UE8M0 FP8 scale data format to ensure compatibility with microscaling data formats.
AI Model Provider
DeepSeek is a Chinese AI company that develops advanced language models optimized for coding, reasoning, and general-purpose tasks.
Their models are designed to be both capable and efficient, offering strong performance across a wide range of applications.
Join innovative teams and creators who are already building the future with intelligent automation. Start free, scale as you grow.