Hacker News: wirybeige

New comment by wirybeige in "Google Chrome update will close the door on ad blockers"

wirybeige — Tue, 16 Jun 2026 14:56:46 +0000

I have noticed it to be slightly worse, mostly on "bad" sites. Just using Brave at this point though.

New comment by wirybeige in "MAI-Code-1-Flash"

wirybeige — Tue, 02 Jun 2026 20:15:17 +0000

Granite is IBM

New comment by wirybeige in "Claude Opus 4.8"

wirybeige — Fri, 29 May 2026 16:49:47 +0000

I presume the providers are the ones giving the info to OpenRouter? I mean, technically it is a mix of fp8 and fp4 (although it is predominately fp4), so I don't think either is inaccurate.

New comment by wirybeige in "Claude Opus 4.8"

wirybeige — Thu, 28 May 2026 20:39:18 +0000

DS4 Pro/Flash were post trained with QAT, so they are already quantized to FP4 for the most part. That's why when downloading the weights, they are much smaller than what their weights at fp8 or fp16 would be. For example, Flash is a 284B model, but its GB size is only ~160GB. OFC maybe DeeppInfra went even further, but there is no proof of that.

New comment by wirybeige in "The Quiet Renovation at Bitwarden"

wirybeige — Mon, 18 May 2026 15:46:58 +0000

Personal anecdote --- Proton Pass very quickly went from worse than Bitwarden to better with more reliable auto-fill.

New comment by wirybeige in "DeepSeek V4 – almost on the frontier"

wirybeige — Sat, 02 May 2026 15:05:58 +0000

These were trained on NVIDIA gpus. It is running inference on Huawei.

New comment by wirybeige in "After dissing Anthropic for limiting Mythos, OpenAI restricts access to Cyber"

wirybeige — Fri, 01 May 2026 15:56:21 +0000

Why would these 3rd-party providers be taking losses? Together, Novita, etc... are not losing money on inference services, they are profiting. You can easily do napkin math with current & last gen Nvidia cards to calculate cost to host/serve these models. I would also doubt that any 1st-party providers like OpenAI and Anthropic lose money on per token billing. There is almost undoubtedly healthy margin being made on that.

New comment by wirybeige in "After dissing Anthropic for limiting Mythos, OpenAI restricts access to Cyber"

wirybeige — Fri, 01 May 2026 15:15:08 +0000

Pricing for DeepSeek V4 flash is $0.14 in/$0.28 out across basically every provider or close to it. It seems most providers just follow the model creator and set their prices to match. V4 pro was set to be $1.74 in/$3.48 out when DeepSeek first announced it; all providers have set their prices to be about that price, & now DeepSeek has set their pricing to $0.435 in/ $0.87 out. I don't know if this is special pricing, or the promise they made for dropping the price when they get more Huawei cards online. It seems that providers like ParaSail, Together, and Novita just set the price when the model comes out and don't compete.

New comment by wirybeige in "Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers"

wirybeige — Sat, 28 Feb 2026 22:32:46 +0000

The vulkan backend for llama.cpp isn't that far behind rocm for pp and tp speeds

New comment by wirybeige in "The first sodium-ion battery EV is a winter range monster"

wirybeige — Sun, 08 Feb 2026 20:25:26 +0000

Sodium batteries don't yet have the scale that lifepo4 batteries have. I'd expect we will see them get cheaper.

New comment by wirybeige in "Linux is good now"

wirybeige — Fri, 02 Jan 2026 02:23:29 +0000

HDR playback in chrome on KDE works as expected from what I can tell. For GNOME 49.2 it does not, it doesn't get the luminance that it should at this time. 49.3 may fix this.

New comment by wirybeige in "OpenGL: Mesh shaders in the current year"

wirybeige — Fri, 10 Oct 2025 14:16:43 +0000

The post links to this: https://github.com/MCRcortex/nvidium

nvidium is using GL_NV_mesh_shader which is only available for nVIDIA cards. This mod is the only game/mod I know of that uses mesh shaders & is OpenGL. & so the new gl extension will let users of other vendors use the mod if it gets updated to use the new extension.

New comment by wirybeige in "Microsoft is plugging more holes that let you use Windows 11 without MS account"

wirybeige — Tue, 07 Oct 2025 14:23:14 +0000

Sorry it doesn't work for you. I don't have that issue. Gnome looks proper in HDR mode for both HDR and SDR content for me.

New comment by wirybeige in "Microsoft is plugging more holes that let you use Windows 11 without MS account"

wirybeige — Tue, 07 Oct 2025 13:19:48 +0000

GNOME has both color management and color representation protocols implemented. HDR works fine on it

New comment by wirybeige in "Ollama Web Search"

wirybeige — Thu, 25 Sep 2025 22:59:30 +0000

Their GUI is closed-source. If someone wants an easy to use & easy to setup app, may as well use LMStudio, which doesn't try to pretend to be OSS. Or use ramalama which is basically just containerizing LLMs and the relevant bits, pretty damn similar to ollama. Or just go back to "basics" and use llama.cpp or vllm.

New comment by wirybeige in "Nvidia buys $5B in Intel"

wirybeige — Thu, 18 Sep 2025 13:51:53 +0000

Xe2 is superior to current AMD integrated already

New comment by wirybeige in "Jujutsu and Radicle"

wirybeige — Thu, 14 Aug 2025 15:02:05 +0000

I think this is the first blog on JJ that has made me want to use it. The flow seems like it could be quite a bit better than git

New comment by wirybeige in "Does Showing Seconds in the System Tray Use More Power?"

wirybeige — Sun, 13 Jul 2025 19:17:31 +0000

It happens on GNOME at the very least, and I would expect every modern platform is the same way.

New comment by wirybeige in "Bazzite would shut down if Fedora goes ahead with removing 32-bit"

wirybeige — Sun, 29 Jun 2025 14:03:57 +0000

The proposal has been retracted, but I fully expect 32-bit to be removed in time.

https://discussion.fedoraproject.org/t/f44-change-proposal-d...

New comment by wirybeige in "The Zed Debugger Is Here"

wirybeige — Thu, 19 Jun 2025 21:53:59 +0000

Setting this to none should remove that button. "features": { "edit_prediction_provider": "none" }