Xiaokang Chen

@PKUCXK

Researcher in Multimodal Team @deepseek_ai Previously Bachelor & Ph.D at Peking University @PKU1898 Opinions are my own.

charlescxk.github.io

Joined June 2022

67 Following

6,468 Followers

Replies Reposts Media

Xiaokang Chen
@PKUCXK
Oct 17
Always improving.😄
DeepSeek
@deepseek_ai
Oct 17
🚀 DeepSeek-V4.1-Flash-1017 is here! Stronger benchmark performance, faster task completion, and shorter outputs than the previous version—a Pareto improvement in performance and cost. Now live on our website, app, and API. 🛠️ Optimized for @DeepSeekHarness, Tencent CodeBuddy,
17
Xiaokang Chen
@PKUCXK
Sep 10
Powered by more extensive RL training, DeepSeek-V4.1-Flash delivers a significant performance boost on visual agent tasks (e.g., agentic visual reasoning) over DeepSeek-V4-Flash-Vision-Exp. 🤓
DeepSeek
@deepseek_ai
Sep 10
🚀 Introducing DeepSeek-V4.1-Flash: smarter, faster, more efficient. 🔹 Introducing the smallest model in our new architecture family, with native visual understanding. 🔹 Designed for greater capability, faster inference, higher throughput, and scaling to larger models. 1/6
10
Xiaokang Chen
@PKUCXK
Aug 21
Welcome to try out this model!😁
DeepSeek
@deepseek_ai
Aug 21
DeepSeek-V4-Flash-Vision-Exp is now live on the DeepSeek API Platform! 🚀 🔹 This experimental multimodal model matches DeepSeek-V4-Flash on text capabilities—including agents, reasoning, and world knowledge. 🔹 On multimodal agent benchmarks, V4-Flash-Vision-Exp makes a major
13
Xiaokang Chen
@PKUCXK
Aug 13
🤓🤓
DeepSeek
@deepseek_ai
Aug 13
We’re launching DeepSeek-V4-Pro today! 🚀 🔷 Major Agent upgrades with strong production gains! 🔷 Flexible reasoning effort for V4-Pro & V4-Flash: low for simple tasks, high for daily Agent workflows, max for complex tasks. 🔷 Native OpenAI Responses API support, optimized for
13
Xiaokang Chen
@PKUCXK
Aug 13
Nice work!😉
DeepSeek
@deepseek_ai
Aug 13
🧩 DeepSeek Harness v0.1 is now available in Developer Preview! 🔹 We’re opening it up to developers building agent harnesses worldwide and open-sourcing the codebase in MIT license. 🔹 Powered by the Cordis meta-framework, DeepSeek Harness is an agent harness built around one
6