Latest DeepSeek News & Updates

See the latest news and media coverage for DeepSeek. We track all announcements, press releases, and industry mentions in real time, all in one place.

Company icon
DeepSeek

Open-source foundation model developer

deepseek.com
Headquarters
Hangzhou, China
Founded year
2023
Company type
Private company
Number of employees
150–200

Latest news about DeepSeek

Company announcements

  • DeepSeek

    DeepSeek releases DeepSeek-V4 Preview

    It features 1M context length, open-sourced models including V4-Pro and V4-Flash, with API access available now.

  • DeepSeek

    DeepSeek introduces DeepSeek-V3.2-Exp experimental model

    Built on V3.1-Terminus, it debuts DeepSeek Sparse Attention for faster, efficient training and inference on long contexts. Now live on App, Web, and API with prices cut by over 50%.

  • DeepSeek

    DeepSeek introduces DeepSeek-V3.2-Exp with DSA

    It builds on V3.1-Terminus and debuts DeepSeek Sparse Attention for longer-context efficiency. API is live on App, Web, and API with prices reduced by over 50%.

  • DeepSeek

    DeepSeek announces V3.1-Terminus improvements

    The update enhances language consistency, addresses user feedback, and strengthens Code Agent and Search Agent performance.

Unlock all announcements with a

Media coverage

  • Techmeme

    An evaluation by NIST's CAISI says DeepSeek V4 Pro lags behind leading US AI models by about eight months and is the most capable Chinese AI model to date

    NIST: An evaluation by NIST's CAISI says DeepSeek V4 Pro lags behind leading US AI models by about eight months and is the most capable...

  • QQ News

    From luxury goods to infrastructure, DeepSeek V4 reconstructs the AI value logic

    Quick read: The DeepSeek V4 series (Pro and Flash) is challenging the leading positions of OpenAI and Anthropic with extremely high parameter efficiency and highly competitive pricing. It not only approaches frontier models in performance, but also makes large-scale, high-frequency Agentic workflows (intelligent agent workflows) possible by significantly reducing inference costs. The emergence of DeepSeek V4 makes people realize that the so-called 'frontier' may not always be synonymous with expensive. If previous models were more like needing expensive...

  • QQ News

    DeepSeek V4 vs Claude Code: Speed Doubled, Costs Plummet, Development Landscape Changes

    Quick Read: Developers found that using DeepSeek V4 instead of Claude Code can achieve several times the response speed and extremely low costs. Although there are still controversies in complex multi-file refactoring and logical depth, its cost-performance ratio is already sufficient to change the cost structure of high-intensity development...

  • MSN

    DeepSeek Partners with Huawei Chips to Drive Domestic AI Ecosystem

    On April 24, DeepSeek released the V4 preview version and announced adaptation to Huawei Ascend chips, attracting attention from industry and capital markets. On April 27, the China Academy of Information and Communications Technology launched V4 domestication adaptation testing, marking the acceleration phase of integration between domestic large models and computing infrastructure. V4 has improvements in architecture optimization, ultra-long context processing, and Agent capabilities, and cooperates with API price reduction strategy ...

Unlock all articles with a

Never miss news about DeepSeek

Track DeepSeek and your other target companies to get real-time alerts and weekly summaries delivered straight to your inbox.

DeepSeek competitors & trending companies

Browse news for competitors to DeepSeek and other trending companies.