frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Ask HN: Local models to support home network infrastructure?

5•DrAwdeOccarim•1d ago
I have had a blast getting Claude Code to manage my home infrastructure. I have been against the cloud forever, so I have had to build a home setup that does a lot of cloud stuff. Like, I run Resillio Sync for all my family iOS photo backups, and a local NAS to host my legally downloaded and owned movies and tv shows, I also use a bunch of raspberry pis, doing things like running local Home Assistant z-wave and zigbee sensors. The router, switches, and APs are all UniFI, same with all the cameras, door bells, and VoIP. Again, all local first (except Talk, obvs).

As you can imagine, maintaining entropy for all these disparate systems takes time, of which I have less now that I have young kids. So when Claude Code was released, I took to it like a fish to water. We mapped my entire network, I created accounts on all the devices so it can SSH into everything and configure everything (including the Ubiquiti Dream Machine Pro!). I have been blow away at how well it troubleshoots and fixes everything.

I have a DGX Spark AI workstation (128gb of memory), and I really want to now hand off the work to a local model, either using Opencode or Claude Code harnesses and simply pointing at a vLLM instantiated model accessable by API (just point Opencode or Claude Code at the local IP and API endpoint).

It works, except I tried Qwen3-coder just now and it's refusing to help due to security concerns. Ugh. I then tried GLM-4.7-Flash, but vLLM doesn't support yet and so before I rebuild (ask Claude Code to rebuild and deploy) to try GLM.4-7-Flash with some other inference provider, does anyone have a model they use for infrastructure maintenance that isn't a little bitch? I will probably eventually go to an abliterated model if none of the open source ones will help.

Comments

actionfromafar•1d ago
There was something on HN recently about how to "trick" the open ones to help.
DrAwdeOccarim•1d ago
OK, I'll look around. Thanks!
jauntywundrkind•16h ago
What had Qwen3-coder rejected? Right now that seems like the strongest recommendation. GLM-4.7-Flash seems very promising but is so new.

Gemma3 is also very good. Nanbeige-4 is supposedly incredibly capable. Both are very small. https://huggingface.co/google/gemma-3-4b-it https://huggingface.co/Nanbeige/Nanbeige4-3B-Thinking-2511

Ideally IMO, you should probably build little tools or a multi-tool for doing the work you want done. Rather than having LLMs having to figure out what needs to be done, doing a more code mode style of development and giving the LLM's the ability to call your tool will be far faster and far more consistent with far lower resources. Tiny models like FunctionGemma will be able to take simple commands and get the work done, very fast, with very little resources. Anthropic wrote this up, citing also CloudFlare calling it Code Mode. https://blog.google/innovation-and-ai/technology/developers-... https://www.anthropic.com/engineering/code-execution-with-mc...

(Note that while Anthropic is suggesting MCP for their "code mode" direction, and while writing MCP's is super easy: writing a cli tool can have just as good as a results! And is often easier for humans to work with!)

Bending Spoons laid off almost everybody at Vimeo yesterday

78•Daemon404•1h ago•38 comments

Ask HN: Do you have any evidence that agentic coding works?

372•terabytest•1d ago•379 comments

Avoid Cerebras if you are a founder

5•remusomega•53m ago•2 comments

Ask HN: Revive a mostly dead Discord server

17•movedx•20h ago•23 comments

Ask HN: COBOL devs, how are AI coding affecting your work?

167•zkid18•2d ago•183 comments

Ask HN: Which common map projections make Greenland look smaller?

17•jimnotgym•23h ago•16 comments

Ask HN: Is retreq / retspec a thing?

2•foobarbecue•4h ago•0 comments

Ask HN: How do you keep system context from rotting over time?

15•kennethops•1d ago•20 comments

Ask HN: Why don't tech companies provide housing?

5•alcasa•5h ago•7 comments

Ask HN: Is it even possible to stop Google Calendar Spam?

4•artur_makly•1h ago•1 comments

Ask HN: How to introduce Claude Code to a team?

8•9dev•1d ago•3 comments

Ask HN: What are the recommender systems papers from 2024-2025?

14•haensi•1d ago•1 comments

Ask HN: What's an API that you wish existed?

9•tornikeo•1d ago•14 comments

Ask HN: Did past "bubbles" have so many people claiming we were in a bubble?

16•bmau5•19h ago•18 comments

Ask HN: Local models to support home network infrastructure?

5•DrAwdeOccarim•1d ago•3 comments

Ask HN: Breaking into tech project management from different field?

4•conner_h5•20h ago•4 comments

Ask HN: How worried should I be about running LLM code on my machine?

9•scoofy•1d ago•4 comments

Ask HN: Should you combine your personal website and blog or keep them separate?

6•nanfinitum•23h ago•3 comments

Ask HN: Clipboard overflows causing system crashes in macOS Tahoe 26.3 beta 2?

8•nhubbard•1d ago•3 comments

Ask HN: How would you design for this scale today?

4•phs318u•1d ago•4 comments

Ask HN: Would you trust a new browser security extension in 2025?

3•linklock•1d ago•8 comments

Ask HN: What non-fiction do you read?

14•yanis_t•1d ago•15 comments

TruCite–an independent verification layer for AI outputs in regulated workflows

3•docmani74•1d ago•0 comments

Ask HN: What should I do with my old laptop in 2026?

5•nanfinitum•1d ago•8 comments

Treating anxiety as a bug in legacy code (engineering approach)

5•bitkin_dev•1d ago•5 comments

AI Californication

6•shoman3003•1d ago•2 comments

Ask HN: Do we need independence and autonomy in Edge-Cloud?

2•Dutchhack•19h ago•3 comments

Ask HN: how to detect teammate vs. enemy in Krunker.io?

2•kracked0x•20h ago•0 comments

Fabric lets me assess online AI from my Unix CLI

2•oldguy101•21h ago•1 comments

Ask HN: Claude Opus performance affected by time of day?

39•scaredreally•5d ago•39 comments