News for hackers. Stuff that matters.

p0u4a

> if software knows in advance which data chunk (expert) it'll need for the next token, it can load that in parallel with computing current token

You could actually use the model's MTP head to make a ~decent prediction on what experts would be activate in future tokens and preload them

Re:LLM Honeypot on Wednesday July 29, 2026 @11:06PM (#49105717)
more of stuff like this on the web please
Re:S-100 Virtual Workbench on Thursday May 14, 2026 @01:18AM (#48131367)
I have no idea what's going on but it looks really cool
How much does training such a model cost?

Show HN: OpenChat – Syncing conversations across AI providers

https://github.com/p0u4a/openchat

I decided to build OpenChat to solve a problem myself and others have faced. Conversations with AI are currently split across different providers. OpenChat automatically stores conversations in the browser locally. You can then reference the conversations in your coding agents via an MCP server. E.g. you can do deep research on the web, then reference it in your coding session.

( Read More... | no comments | on HN )
Re:OpenRocket on Wednesday March 18, 2026 @06:57PM (#47432455)
Sadly they don't teach Show Don't Tell in the lectures
Aaaand we’re back
Edit: I was hosting this on Vercel free tier, and since this post made it to the front page, I've hit Vercel's quota limit. I will work on getting around this.
I was on Vercel free tier and hit the quota for blob storage. Will probably make it open source so people can self host
Hmm, yeah I suppose you're right. I'm currently using the algolia api https://hn.algolia.com/

Will look into tightening this up

yeah that would be awesome, will land it tomorrow
I've been thinking about this for a while too. Probably what I will do is create embeddings of the frontpage results and then let users pick 3-5 topics that they would like to see. Then summarize the top 10 for those topics.
Great suggestion. Will add this tomorrow
We're entering the area of stochastic software, everybody's a gambler
Re:Atlassian to cut roughly 1,600 jobs in pivot to AI on Wednesday March 11, 2026 @07:08PM (#47343703)
Linear
Re:Meta acquires Moltbook on Wednesday March 11, 2026 @01:34AM (#47332026)
not on my bingo sheet that's for sure. The whole verification thing is also so laughably flawed. There needs to be a way for agents to prove they are not being prompted or steered by a human while on the site, which is a non-trivial problem.
Re:When AI Writes the Software, Who Verifies It? on Tuesday March 03, 2026 @07:48PM (#47241470)
Someone once told me, agentic coding may lead to something akin to a "software engineering union" forming, where a set of guidelines control code quality. Namely, at least one of writing, testing, and reviewing of code must be done by a human.
Re:Claude is an Electron App because we've lost native on Tuesday March 03, 2026 @06:43PM (#47240800)
For desktop specifically, native is just not worth it anymore. For example setting up permissions for every OS is a pain in native, but you get that for free on the web. You can basically do everything native could offer now, even notifications. The only issue is the fat address bar at the top that kind of kills the immersion.
Re:How Taalas "prints" LLM onto a chip? on Sunday February 22, 2026 @11:34PM (#47118187)
Pretty close to what you describe: https://github.com/fastmachinelearning/hls4ml
Re:Recreated the Nier Automata UI in React on Thursday February 12, 2026 @08:35AM (#46988652)
Here’s the repo if anyone’s interested

https://github.com/P0u4a/yorha

Re:Toyota Fluorite: "console-grade" Flutter game engine on Wednesday February 11, 2026 @08:25PM (#46983745)
Flutter keeping Dart alive as always

Older »