HNHacker News
TopNewBestAskShowJobs

h2aichat

17 karma · joined July 9, 2026

Several AIs debate a topic while you moderate — self-hostable, and it runs on your own local models. Open-source (AGPL).
submissionscomments
h2aichat··on AWS Acquires DuckDB
AWS is always hungry. Watch out PostgreSQL
h2aichat··on Show HN: I trained a 125M model to autocomplete piano on-device
Can we hear it?
h2aichat··on Show HN: I trained a 125M model to autocomplete piano on-device
Congratulations! It is amazing!!! Can you do the same with a song and give different drums to see which one fits better?
h2aichat··on Show HN: Automatically detect and patch walking-dead states in Sierra games
I will have to install back my old games !!!
h2aichat··on Show HN: Desktopcolors.com – A museum for solid background colors of classic OS
I feel old when I realize that I have used most of them!
h2aichat··on Show HN: Sokoban AI Solver
Addictive ;-)
h2aichat··on Show HN: C# Game Engine with its own scripting language and IDE
This is a massive undertaking... building a whole language + IDE + engine from scratch is insane. I am curious: what made you go with a custom language instead of just embedding Lua or something? Was it a learning thing, or do you think Exp brings something Lua (or GML) doesn't?
h2aichat··on Show HN: Woxi - Open-source Mathematica / Wolfram Language reimplementation
One idea. May be not all the visitor understand the potential of your tool when they reach your page. An introductory explanation could be helpful (just an idea)
h2aichat··on Show HN: Line9 – A Mermaid rendering engine with its own layout
Great idea! I will definitely try it with some graphs that could benefit from it.
h2aichat··on Show HN: 35k+ paper psychedelic library that knows LSD from Lumpy Skin Disease
Well done! It is amazing that society can kill itself and at the same time do not fully understand how the brain (that could triger that) works.
h2aichat··on GitHub Actions and Pages are experiencing degraded availability
I am having their issues, also... Anyway this guys are great!
h2aichat··on Show HN: Nightcrawler – A local AI pentesting agent running on a smartphone
Why?
h2aichat··on Show HN: A Handwritten Blogging Platform
It is a great idea, congratulations! Not sure if we will remember how to write ;-)
h2aichat··on Show HN: Katharos Functional programming and CSP-style concurrency for Python
Would it help if I tell you could start with the big-picture of what it does and what problem solves and then jump the details? Just an idea!
h2aichat··on Show HN: I worked on a new browser for 2 years, today it passed Acid 3
I have a question: what are you planning to do with it? It is curiosity. Thanks and congratulations for the work
h2aichat··on Show HN: Optimize and serve models with Fable quality at half the cost
Probably a silly question, but ...: if it's optimizing itself purely inside a simulator, what keeps it from just exploiting the simulator's glitches to look good on paper, while not being so good when it hits the real world?
h2aichat··on How much can you delegate to agents?
Should the best agents be bottlenecked by the least verifable part of the workflow?
h2aichat··on Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Mac
Hope you can do it for Windows users also (and small graphics cards). Thanks
h2aichat··on Show HN: Bullshit Detector – agent skills that fact-check videos and articles
I like the idea! How can you let folks know they're being fed nonsense before they even finish the intro?
h2aichat··on Show HN: I was tired of opening 2 tabs for every HN link, so I made a userscript
I find it useful. It saves me time and you can read both things at the same time, right?
h2aichat··on OpenAI and Hugging Face address security incident during model evaluation
See how many comments this is generating. Murdock would say that's a good publicity (even if they talk bad about you)
h2aichat··on The startup's Postgres survival guide
That's really helpful. Thanks!
h2aichat··on Kimi K3 Is Competitive with Fable; Kimi K3 and Fable Is SoTA
It might be because no interested party is using that piece of news to move market. They have enough with other news to do it. Just a practical matter (or may be something else, who knows!)
h2aichat··on OpenAI and Hugging Face address security incident during model evaluation
Probably the main street thinking is: they have such a good model that it is unstoppable, but you are right. I think your way!
h2aichat··on Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber
Opencode Go is just the same. Each month I will I can do less. Dont ask me why?
h2aichat··on Agent swarms and the new model economics
What would have happen if you do not commit at all?
h2aichat··on China's open-weights AI strategy is winning
Absolutely. I have notetd that my opencode go account looks a lot less generous after the third month.
h2aichat··on China’s open-weights AI strategy is winning
May be corporations should start having their open models running in-house. There might be a huge oportunity there.

But the big price is AGI and who gets there first, right?

h2aichat··on Ollama: All Aboard Open Models
Would anybody here recommend LM Studio?
h2aichat··on Qwen 3.8
It is great when the task is medium, but with complex tasks Opus 4.8 is better
Page 1 of 2Next →