Kimi-VL Technical Report

Research scientist at Moonshot AI studying scaling in reinforcement learning and large language models.

Xinxing Zu is a research scientist at Moonshot AI. His public profile notes a PhD in computer science from New York University, prior work as an AI scientist at Amazon, and research interests in AI agents, reasoning, and multimodality.

Xingcheng Yao is a research scientist at Moonshot AI. His public profile notes prior work as a research engineer at Tencent AI Lab, a PhD in computer science from the University of Southern California, and research interests spanning NLP, multimodal systems, and AI agents.

PhD student at Renmin University of China whose homepage highlights multimodal foundation models and video understanding, with public projects and papers including DeepSeek-VL and Kimi-VL.

Researcher at Moonshot AI working on multimodal large language models; previously a key member of Alibaba's Qwen team and author of work including Kimi-VL, DeepSeek-VL, and Qwen technical reports.

Yulun Du is a Moonshot AI-affiliated researcher. Public profiles also show prior work and study at Carnegie Mellon University, including a Master of Language Technologies completed in 2020.

Researcher and engineer focused on reinforcement learning and embodied intelligence; his public profile lists work spanning Huawei Noah's Ark Lab, Momenta, Moonshot AI, and XVI Robotics, and he is credited on Moonshot AI technical reports.

Research scientist at Moonshot AI working on foundation models, multimodal large language models, and agents; previously worked at Huawei Noah's Ark Lab and studied at the Chinese University of Hong Kong.

Research scientist at Moonshot AI focused on large multimodal models and large language model post-training.

Dongliang Wang is a research scientist at Moonshot AI whose public profiles highlight multimodal large language models. His homepage also notes earlier PhD work at Shanghai AI Lab and Shanghai Jiao Tong University.

Huabin Zheng is a research scientist at Moonshot AI. His homepage says he works on large language models, multi-agent systems, code generation, and game agents.

Assistant Professor of Computer Science at the University of Hong Kong and director of XLANG Lab, focusing on natural language processing and embodied AI agents.

Pengyu Cheng is a research scientist at Moonshot AI. His homepage says he works on multimodal large language models, reinforcement learning, and agents, and previously interned at Microsoft Research.

Researcher at Moonshot AI. Public profile notes computer science and engineering study at HKUST from 2021 to 2025.

Research Scientist at Moonshot AI whose public work focuses on large language models, multimodal models, and embodied AI; he previously earned a PhD from Zhejiang University and was a visiting student at Oxford.

Research scientist and writer behind Scientific Spaces whose public profile lists work on large language models and service on the Kimi team at Moonshot AI.

Researcher working on multimodal learning and vision-language systems, with public academic work on visual question answering and related topics.

Research scientist at Moonshot AI focused on multimodal large language models.

Associate research scientist at Moonshot AI based in Beijing, China; previously worked as a postdoctoral researcher.

Machine learning researcher at Moonshot AI and incoming assistant professor at Shanghai Jiao Tong University.

Multimodal and omni-model engineer whose public profile lists Moonshot AI experience and Kimi-VL among recent projects.

Technical staff member at Moonshot AI whose public profile highlights work on web and app agents, multimodal systems, reinforcement learning, and LLMs.

Dikang Du is a research scientist at Moonshot AI. His homepage says he received a Ph.D. from Cornell University and works on natural language processing, machine learning, and multimodal learning.

Technical staff member at Moonshot AI working on general AI agents, reinforcement learning, and multimodal foundation models.