DeepSeek-V3 Technical Report

Yukun Zha is a research scientist at DeepSeek. His public profile lists research interests in machine learning, reinforcement learning, and efficient inference, alongside a PhD in computer science from Dartmouth College.

Researcher at DeepSeek whose public homepage describes work on DeepSeek R1, V1, V2, V3, Math, Coder, and mixture-of-experts systems.

Liyue Zhang is a research scientist at DeepSeek whose public profiles highlight multimodal models, reasoning, and AI agents.

Researcher at DeepSeek AI focusing on multimodal large language models and multimodal understanding.

Research scientist at DeepSeek AI. Previously completed a PhD at Carnegie Mellon University.

Wangding Zeng is a researcher at DeepSeek. His OpenReview profile also lists graduate and undergraduate study at Beijing University of Posts and Telecommunications.

Researcher at DeepSeek focused on large language models, reinforcement learning, multimodal large language models, and generative AI.

MPhil student in the Department of Computer Science and Engineering at HKUST interested in LLM agents and multimodal models; previously interned with the DeepSeek LLM Alignment Team.

Researcher at DeepSeek focused on multimodal large language models, world models, reinforcement learning, and embodied AI.

Research scientist at DeepSeek interested in reinforcement learning, large language models, and reasoning.

Han Bao is a research scientist at DeepSeek AI. His homepage says he works on machine learning theory, including out-of-distribution generalization, robust machine learning, self-supervised learning, and learning-based control.

PhD student at Shanghai Jiao Tong University advised by Prof. Xueqi Cheng. Her research interests include retrieval and agents, multimodal large language models, and biomedical large language models, and she is currently an intern at DeepSeek.

Researcher at DeepSeek AI and incoming Ph.D. student at the University of Hong Kong working on efficient and trustworthy LLMs, multimodal large language models, and reasoning.

Research scientist at DeepSeek interested in post-training, tool use, and in-context learning.

Research scientist at DeepSeek AI focused on multimodal large language models, reinforcement learning, and efficient post-training.

Researcher in natural language processing and pre-trained language models, with public academic work from Tsinghua University and related NLP publications.

DeepSeek researcher focused on NLP, code intelligence, and LLM reasoning, with public work spanning DeepSeek-Coder, DeepSeekMath, DeepSeek-V2, DeepSeek-V3, and DeepSeek-R1.

Researcher at DeepSeek AI working on decision-making and post-training for large language models.

Research scientist at DeepSeek interested in large language models, reinforcement learning, robot learning, and machine learning.

DeepSeek engineer and co-author of the DeepSeek-V3, DeepSeek-V2, and DeepSeek LLM technical reports.

Research scientist at DeepSeek.

First-year Ph.D. student in Information Engineering at the Chinese University of Hong Kong advised by Dahua Lin and Jifeng Dai. His research interests include understanding and building intelligence in multimodal large language models.

Research scientist at DeepSeek working on post-training and reinforcement learning, with interests in systems and AI for science.

Research scientist at DeepSeek with public GitHub projects on large language models and AI systems.