Kwai Keye-VL-2.0：开源MoE多模态模型，支持256K长视频理解
Kwai Keye-VL-2.0 Technical Report
长视频理解和智能体场景的开发者终于有了一个开源且高效的MoE模型——Keye-VL-2.0仅激活3B参数就能处理256K上下文，做视频分析或智能体应用的团队可以直接下载权重试试。
快手发布Keye-VL-2.0-30B-A3B，一个基于MoE架构的开源多模态基础模型，专为长视频理解和智能体场景设计。
该模型首次将DeepSeek Sparse Attention适配到GQA架构，实现无损256K上下文处理，能捕捉关键帧和长程时序依赖。通过跨模态多教师策略蒸馏（MOPD）和上下文/视频强化学习，解决了多任务对齐中的灾难性遗忘问题，仅激活3B参数即可在代码、工具和搜索场景中实现智能体协作与多模态自我纠正。
在视频理解、时序定位、推理、STEM和智能体基准测试中，Keye-VL-2.0在相似规模模型中达到最优性能，尤其在TimeLens细粒度时序定位和Video-MME-v2长视频理解上表现突出。
模型权重已开源。
Kwai Keye-VL-2.0 Technical Report
We introduce Kwai Keye-VL-2.0-30B-A3B, an open-source Mixture-of-Experts (MoE) multimodal foundation model designed to advance long-video understanding and agentic intelligence. To address the challenges of ultra-long contexts, information redundancy, and prohibitive computational costs inherent in hour-level videos, Keye-VL-2.0 is the first to adapt DeepSeek Sparse Attention (DSA) to GQA-based multimodal architectures, enabling lossless 256K context processing while capturing critical frames and long-range temporal dependencies. This architecture is underpinned by a highly optimized training and inference infrastructure, including scalable video I/O, heterogeneous ViT-LM parallelism, and custom DSA kernels that significantly maximize throughput and minimize computational overhead. Furthermore, to overcome the algorithmic dilemma of catastrophic forgetting during multi-task alignment, we introduce Cross-Modal Multi-Teacher On-Policy Distillation (MOPD) paired with Context-RL and Video-RL. By distilling dense token-level teacher feedback from on-policy rollouts back into the MoE backbone, which activates only 3B parameters, Keye-VL-2.0 natively empowers advanced agent collaboration across Code, Tool, and Search scenarios with multimodal self-correction. Extensive evaluations across video understanding, temporal grounding, reasoning, STEM, and agent benchmarks demonstrate that Keye-VL-2.0-30B-A3B achieves state-of-the-art performance among models of similar scale, particularly excelling in fine-grained temporal localization on TimeLens and long-video comprehension on Video-MME-v2 and LongVideoBench. We release our model checkpoints to accelerate community progress toward scalable and robust multimodal agentic applications.
This report is categorized under AITOP's "Models" desk. Key entities and technical topics under continuous tracking include 开源/仓库, MoE, 长视频理解, 智能体.
Primary source verification: published directly by arXiv: DeepSeek with first-party authority.
