Home|News|Photo|Opinions|CCYL|Fun|Finance|Military|Sports|Employment|University|Travel|Discovery|Video|Games|Autos|Youth Inspring Stories
DeepSeek-V3.2-Exp Enhances Training and Inference Efficiency

Time:2025-09-30 14:52:00 Source: Youth.cn China Youth International

【Tech 24H】DeepSeek has released the DeepSeek-V3.2-Exp model, an experimental version serving as an intermediate step toward a new-generation architecture. Building upon V3.1-Terminus, V3.2-Exp introduces DeepSeek Sparse Attention, a sparse attention mechanism designed to explore and optimize training and inference efficiency for long-text processing. The model has now been updated across DeepSeek's official app, web platform, and mini-program. Additionally, the API pricing has been significantly reduced, cutting costs for developers by over 50% under the new pricing policy. The DeepSeek-V3.2-Exp model now made its open-source on Huggingface and ModelScope.

Editor:Hou Qianqian
Online Dictionary:

About UsContact UsAdvertiseJobsIllegal Information Reporting Send qnb to 10658000 to order Mobile China Youthz

Organized by CCYL and Network Film & TV center of CCYL Copyright@China Youth International. All rights reserved.
信息网络传播视听节目许可证0105108号 京ICP备13016345号-8   京公网安备 11010102004843号