我觉得国内应该全面禁止AR。对于一群做着中国梦的人来说，“增强现实”是致命的 （手动滑稽

I Have a Dream. That One Day, the world’s most advanced intelligence — whether it’s called GPT, Fable, DeepSeek, or something else Just Does Not Matter~ — will be just like electricity and water supplt today. Cheap. Stable. High‑quality. It will power all kinds of AI software, the way electricity powers all of akindsppliances. We’re still living in the era when the light bulb was just invented. There will be countless AI applications in the future. That’s why token production and token transportation — as the underlying infrastructure — will be the most important infrastructure build of the next 10 to 30 years. This is an industry‑wide upgrade. A benefit to all of society. And this is exactly why I strongly oppose closed‑source models. #AGIForEveryone How can we let infrastructure — something as essential as water and electricity — become a luxury for the few? How can we expect people to pay such exorbitant bills? #路长而歧行则将至 #我心匪石不可转也 #AI #Infrastructure #OpenSource #TokenEconomy

Thanks for all the feedback. GLM-5.2 will begin rolling out to all Coding Plan users in 3 hours.

Finally, I found the reason for the usage limit being reset this afternoon. It's the best news of the day.

In OpenCode Go's statistics, the one with the highest usage turns out to be DS V4 Flash instead of Pro — interesting. 🧐

Kurnal want Kurnal got analyzing Intel data center GPU card PCB work with Void-tech lab XD

Today, when calling the Kimi Code API via the pi agent core, it keeps returning errors. Did you break something again? 400 {"error":{"type":"invalid_request_error","message":"The request was rejected because it was considered high risk"},"type":"error"} @Kimi_Moonshot @zxytim @Young_AGI

Long semi-short SAAS unwinding for the second day Thanks everyone🙏

I follow lots of US AI folks and know them by name and have a good idea of their achievements & sometimes even personalities. I found that Chinese researchers were kind of a blob, even to me. So I started working on some projects to make the people driving China's AI labs more visible. (Or at least that's where I started.) Their achievements, visions and yes even personalities are having an outsized impact in this crazy AI age. Here's a sneak peek, lol. Do you love how absolutely nerdy this is?

I’ve been relying on Kimi’s web-based professional databases for equity research and fact-checking for a while now, and the experience has been rock solid. The catch? Everything was manual — you had to click through the web UI. The new kimi-datasource plugin solves exactly that: all those tedious web tasks can now be automated and integrated directly into my agent workflow. Huge win. kimi.com/code/docs/en/kimi-c…

Summarization (you can get all of them on the internet, not insider information): - M3 - No more M2.x, e.g., no M2.8 - MoE, 1M context - Larger than M2 (200~B model) - MSA, aka MiniMax Sparse Attention architecture - For MSA, 9.7x prefill speedup, 15.6x decoding speedup - Will be released in a few days - Will be open-sourced

Huawei just confirmed the Mate 90 series is packing the Kirin 2026 — 3.1 GHz, 238M transistors/mm², all built on their τ-law design. Now we just need TechInsights to rip one open and tell us what's actually inside!

LFG! MiniMax M3 will be released soon!

Deepseek permanently sets the price of V4-Pro to one quarter of the original price.

The Involuntary Pomodoro in the AI Age.

IF oil > 100 AND bond > 5%: TRUMP.TACO() ---- It's time to TACO.

Come try out the incredible work from our genius multimodal colleagues! 🐳👀 The little whale can now see (in grayscale testing)~ ✨

DeepSeek has extended the 75% off discount on the Pro model until the end of May.

After Russia imposed temporary export restrictions on helium, helium prices in China surged. Per March 2026 customs data, 65% of China's helium imports came from Russia.

Xiaomi unveiled its full-sized humanoid robot. mi.com/cyberone