Hacker News: linzhangrun

Maextro V800: brake pedals fail; videos pulled, no recall, astroturfing alleged

linzhangrun — Sun, 11 Oct 2026 02:29:20 +0000

Article URL: https://chinaevhome.com/2026/10/09/huaweibacked-maextro-v800-braketest-videos-pulled-from-platforms/

Comments URL: https://news.ycombinator.com/item?id=50039243

Points: 4

# Comments: 0

New comment by linzhangrun in "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"

linzhangrun — Fri, 09 Oct 2026 01:21:43 +0000

There are too many commoditized models to count: GLM5.3 Flash, Kimi K2.8, Mimo V2.6, MiniMax M3.1...

Luainstaller: "Pyinstaller" for Lua [video]

linzhangrun — Fri, 02 Oct 2026 01:35:28 +0000

Article URL: https://www.youtube.com/watch?v=ctYfIRZeuYc

Comments URL: https://news.ycombinator.com/item?id=49928998

Points: 2

# Comments: 0

ZCode, embroiled in a controversy over stealing user code, is now open source

linzhangrun — Mon, 21 Sep 2026 02:37:38 +0000

Article URL: https://github.com/zai-org/ZCode/blob/main/README.en.md

Comments URL: https://news.ycombinator.com/item?id=49782440

Points: 5

# Comments: 0

New comment by linzhangrun in "How come AI-related posts get so many points on HN?"

linzhangrun — Sat, 19 Sep 2026 08:00:31 +0000

Just like AI-related projects on GitHub always get an entirely different order of magnitude of stars. This is what people are focused on, outshines anything else.

New comment by linzhangrun in "Human brain is two separate organs, Stanford Medicine-led research finds"

linzhangrun — Sat, 19 Sep 2026 07:58:53 +0000

Actually, the mind should only exist in the forebrain:0

New comment by linzhangrun in "Android 17 is the first since 3.x to add new APIs without releasing to the AOSP"

linzhangrun — Sat, 19 Sep 2026 07:55:40 +0000

How surprising.

New comment by linzhangrun in "Show HN: What if the speed of light was 5 km/h?"

linzhangrun — Thu, 10 Sep 2026 08:17:08 +0000

The speed of light, wave-particle duality, and so on, always make me feel that the universe also has a "computing power"

New comment by linzhangrun in "DeepSeek v4.1 Flash"

linzhangrun — Thu, 10 Sep 2026 08:15:48 +0000

They say v4.1flash is so strong that they'll route API calls to v4pro to v4.1flash, lol

super fast true

CodeX v0.148.0 cannot use GPT-5.6 Sol

linzhangrun — Wed, 19 Aug 2026 07:15:26 +0000

Article URL: https://github.com/openai/codex/issues/39397

Comments URL: https://news.ycombinator.com/item?id=49358030

Points: 1

# Comments: 1

New comment by linzhangrun in "Qwen3.8 27B scores 52 on Artificial Analysis"

linzhangrun — Tue, 18 Aug 2026 08:32:48 +0000

Under the same computing power, the improvement in LLM intelligence and the improvement in the upper limit of LLM intelligence are equally astonishing. At least in coding, the best local models that can run smoothly on a DGX Spark are now less than one year behind the strongest SOTA models in capability (I measured around 60 tok/s).

New comment by linzhangrun in "DeepSeek Harness developer preview"

linzhangrun — Fri, 14 Aug 2026 01:41:05 +0000

Why is the Chinese text missing in many places?

New comment by linzhangrun in "Gemini 3.7 Flash"

linzhangrun — Fri, 14 Aug 2026 01:10:34 +0000

Feels like Gemini Pro will arrive directly as Gemini 4 Pro

New comment by linzhangrun in "ChatGPT Desktop (Codex Desktop) for Linux"

linzhangrun — Thu, 13 Aug 2026 09:55:59 +0000

I don’t think GPT-5.6 would make such a basic mistake when reviewing CodeX itself.

New comment by linzhangrun in "Why tiny JPEGs look different in Chrome"

linzhangrun — Thu, 13 Aug 2026 09:49:26 +0000

Could this be related to the GPU? I remember that sites like Reddit use subtle differences in GPU rendering to fingerprint devices.

New comment by linzhangrun in "ChatGPT Desktop (Codex Desktop) for Linux"

linzhangrun — Thu, 13 Aug 2026 09:40:03 +0000

I tried it yesterday. The experience seems smoother than on Windows. I hope Computer Use will be added soon. Also, when will OpenAI fix the issue where Xhigh and Ultra are both translated as “极高” in Chinese? It’s been this way for quite some time. As far as I know, the proportion of Chinese employees at these Silicon Valley AI companies is quite high.

New comment by linzhangrun in "H3-metal – Native MiniMax-H3 inference for Apple Silicon"

linzhangrun — Tue, 11 Aug 2026 16:51:27 +0000

20%

New comment by linzhangrun in "H3-metal – Native MiniMax-H3 inference for Apple Silicon"

linzhangrun — Tue, 11 Aug 2026 08:53:19 +0000

First batch of quick test results: approximately 1/5 speed improvement

New comment by linzhangrun in "H3-metal – Native MiniMax-H3 inference for Apple Silicon"

linzhangrun — Tue, 11 Aug 2026 06:39:08 +0000

On my 128GB M4 Max Mac Studio, generating a 15s 480p video with MiniMax H3 in ComfyUI takes an hour and a half.

Put Codex to work on deploying it now, hoping the speed can improve quite a lot :-) Thanks anyway

New comment by linzhangrun in "AMD acquires Taalas to boost inference performance by etching models in silicon"

linzhangrun — Fri, 07 Aug 2026 01:07:23 +0000

Thinking that five or six years from now, Fable-level intelligence could be provided at 100x the current speed... makes me feel lost. I cannot imagine what the future will look like.