DeepSeek V4 Pro (0813)

系统要求

不同量化方法和上下文大小的显存要求

关于 DeepSeek V4 Pro (0813)

DeepSeek V4 Pro 0813 代表了 DeepSeek 旗舰级混合专家模型（MoE）的正式发布版本。它具备前沿的推理能力、复杂的软件工程能力以及扩展的上下文分析能力。

技术规格

资源

关于 DeepSeek V4

DeepSeek-V4 is DeepSeek's latest generation of highly efficient Mixture-of-Experts language models, featuring a novel hybrid attention architecture combining Compressed Sparse Attention (CSA) and Heavily Compressed Attention (HCA) that dramatically improves long-context efficiency. Pre-trained on 32T+ tokens with a comprehensive post-training pipeline including domain-specific expert cultivation and unified model consolidation. Both V4-Pro and V4-Flash support 1M context length as standard, with three reasoning effort modes (Non-think, Think High, Think Max). Released open-source under MIT license on April 24, 2026.

其他 DeepSeek V4 模型

• DeepSeek-V4-Flash
• DeepSeek V4 Flash (0731)
• DeepSeek-V4-Flash-Vision-Exp
• DeepSeek-V4-Pro