Tiezhen WANG(@Xianbao_QIAN):Kimi or the entire Chinese open source ecosystem is marching towards linear / hybrid attention. K3, despite being way larger to K2, is 2.5x more efficient. Model arch improvement and intelligent efficiency is the way moving forward to sustain scaling law while still being affordable and eco-friendly.

Tiezhen WANG

@Xianbao_QIAN

Helping ecosystem to grow on vLLM, ex-Head of APAC ecosystem @huggingface. Ex-Googler on TFLite/micro. Ideas on my own. Interested in future tech. DM open

Joined November 2022

2.8K Following 12.2K Followers

Tiezhen WANG@Xianbao_QIAN

2026.07.28 00:49

Kimi or the entire Chinese open source ecosystem is marching towards linear / hybrid attention. K3, despite being way larger to K2, is 2.5x more efficient. Model arch improvement and intelligent efficiency is the way moving forward to sustain scaling law while still being affordable and eco-friendly.