Yiyang Zhu

抖音@创野小朱，半年10W+粉丝~ 科技科普/AI/网络安全
100,000 + Followers in China
JHU - Security Engineer - Influencer

[Yiyang Zhu @cy_xiaozhu]

这是我在推特的第一条帖子，也是我海外社交媒体的第一条帖子hh。从今天开始我决定在推特上多发动态，希望大家多多关注！

[Yiyang Zhu @cy_xiaozhu]

https://t.co/BsY84N4CAP

[Yiyang Zhu @cy_xiaozhu]

真人版豆包老师，好好看哈哈哈哈哈

[Yiyang Zhu @cy_xiaozhu]

已冲，勿辜负

[Yiyang Zhu @cy_xiaozhu]

你也想吃？

[Yiyang Zhu @cy_xiaozhu]

@zarazhangrui Correct, I usually use AI to produce quality input for me🤣

[Yiyang Zhu @cy_xiaozhu]

@dessaigne Then I won't be anxious until I turn 37🤣

[Yiyang Zhu @cy_xiaozhu]

上个月有幸在Google见过Zara，很厉害！也改变了我对运营X的看法：X不同于抖音，更像是你要主动去融入一个圈子👏

[Zara Zhang @zarazhangrui]

I just reached 70k followers on X 🙏🏼

X is where I learn in public & build in public. I've learned so much and met so many wonder people here

If you're curious about how I grew my followers while being completely authentic, read my article below

[Yiyang Zhu @cy_xiaozhu]

在Trae上免费试用了GLM5.2，确实不错，能力应该可以持平Opus4.6。但依旧会骗我说自己已经实现完了，其实没有🤣

[Yiyang Zhu @cy_xiaozhu]

@patrickstone222 亏的也远超同龄人

[Yiyang Zhu @cy_xiaozhu]

@elonmusk @yunta_tsai kid is like a 2-b model, no CoT and too much hallucination🤣

[Yiyang Zhu @cy_xiaozhu]

来咨询一下大家是怎么管理自己已上线的产品的，每一个产品都需要单独租一台服务器再配一套域名么？有没有��价比又高，速度又快的工作流？

[Yiyang Zhu @cy_xiaozhu]

@aronhouyu “请把项目中所有vibe的痕迹都清除，但不要影响主要功能！”

[Yiyang Zhu @cy_xiaozhu]

确实

[Yiyang Zhu @cy_xiaozhu]

@0xajc 散是满天翔

[Yiyang Zhu @cy_xiaozhu]

@0xajc @LMDC5288 逆天

[Yiyang Zhu @cy_xiaozhu]

@litekakacom 特指架构不那么复杂的软件吧。一旦涉及到有高性能要求的软件或者涉及硬件，就不会那么卷了～

[Yiyang Zhu @cy_xiaozhu]

A 256GB unified memory Mac Studio (M3 Ultra configs) costs ~$5,600–$8,000+ from Apple depending on storage.

Electricity for running the 2-bit GLM-5.2 (239GB model): typically 100-300W during inference. At $0.15/kWh that's just **2-5¢ per hour**.

Super efficient once you own the hardware. Unsloth + llama.cpp or their Studio makes it straightforward on Apple Silicon.

--by grok

[Unsloth AI @UnslothAI]

GLM-5.2 can now be run locally!🔥

The 2-bit model retains ~82% accuracy after we shrunk it from 1.51TB to 238GB (-84% size).

Run on a 256GB Mac or RAM/VRAM setups.

GLM-5.2 is the strongest open model to date.

Guide: https://t.co/bI7FeeKHDd
GGUF: https://t.co/BMkxswdj5N

[Yiyang Zhu @cy_xiaozhu]

新一轮的供应链投毒要来了？

[莱特卡卡 @litekakacom]

太快了，Skill还没搞明白多少，loops集合站都已经���来了，以后都不用装skill了，直接装loops就好了：https://t.co/V28t4dogph。

这个站已经收集了40个loops，你可以复制用于coding agents的闭环工作流。

每个闭环都包含触发器（triggers）、反馈门（feedback gates）和退出条件（exit conditions），以便代理能够自我调整节奏，直到任务完成。

[Yiyang Zhu @cy_xiaozhu]

@litekakacom 新一轮的供应链攻击要来了🤣🤡

[Marc Andreessen 🇺🇸 @pmarca]

Interesting. (Founder of https://t.co/PY9jRPdR6L, creator of GLM AI models.)