DeepSeek-V2

Headlines · 2

1. 2d
   CPU-Only Reproduction of Kimi K3 Gated MLA

   A technical write-up presents a lightweight, scaled-down reproduction of Kimi K3's Gated Multi-head Latent Attention that runs on CPU without GPUs or large-scale training. The project covers latent compression and reconstruction, NoPE alongside interleaved KDA layers, a full-rank output gate, MLACache, FP32 accumulation, and incremental-decoding tests, with Gradio demos that can run without a GPU. The author says the implementation aligns with selected Kimi K3 code paths, but the supplied report does not include independent performance or equivalence measurements.

   掘金 · 人工智能 · 🔥 2
2. 2d
   Commentary Highlights Younger Leaders in China’s AI Research

   A commentary from the WeChat account 人人都是产品经理, credited to 怪哥, says researchers born in the 1990s and 2000s are leading important AI work at major Chinese technology companies. It cites Yao Shunyu at Tencent’s Hunyuan, Sun Tianxiang at Baidu, Luo Fuli at Xiaomi’s MiMo, Zeng Yan at ByteDance’s Seedance 2.0, Zeng Aohan at Zhipu AI’s GLM pretraining and infrastructure, and Zhao Pengyu at MiniMax. The commentary argues that scarce AI talent and rapid model development are weakening traditional seniority-based promotion rules. The personnel details and interpretation are not independently corroborated in the supplied material.

   虎嗅 AI · 🔥 3

Experience and discussion from the community

Share my DeepSeek-V2 experienceAsk about DeepSeek-V2