News for hackers. Stuff that matters.

Juvination

Re:Apple ML Research: Muse Spark Open with MLX on Mac Studio on Friday October 16, 2026 @11:58AM (#50120044)
Exactly why does Apple need to do any model training/development? They are out here selling the shovels.
I came in here to suggest this channel, it's an absolute blast hearing him talk about the N64 hardware with such passion and interest in optimization.
Re:Beauty in DVD Menus on Thursday October 08, 2026 @10:57AM (#50006596)
I was just thinking back to how fun the DVD menus were as a kid. Sometimes you'd even have games on them, not that they played very well at all or I never used the right remote. I wish Plex had a feature of dropping an iso, and being able to get the whole dvd/blu-ray menu experience.
Re:EmbeddingGemma 2 on Tuesday October 06, 2026 @05:47PM (#49984594)
So what are some use cases people have found for running these sized multimodals on their device? What is it accurate on, and what is the hallucination rate like?
Re:RuneScape's Position on Gen AI on Sunday October 04, 2026 @12:57PM (#49955613)
One of the most wild things is how actual lazy this image gen was. Hands have been solved on all frontier models, so what did their partner even use?
What are the odds of there being smaller exoplanets that are effectively being outshone by the others? Since each one these is more massive than Jupitar.
Re:Sites in ChatGPT on Friday October 02, 2026 @02:34PM (#49936950)
I made it through the 6th puzzle, it was a lot of fun. I basically just stayed in 3 dimensions until I had to move up and then went back to 3.
Re:Pi 1.0 on Thursday October 01, 2026 @04:24PM (#49926639)
So Pi is a lot of fun. I’ve been working on Volt, a fork of Pi, mostly out of my frustrations with remote development. I expanded out the existing RPC and built a mobile app around it.

It kind of just gets with how you creative you want to be about it.

Re:Strands Harness on Wednesday September 23, 2026 @11:41AM (#49817799)
For what it's worth I've been using Anthropic models on Pi for months now with no issues. It's not recommended since it breaks TOS but you can do it.
Re:Claude Fable 5.1 and Claude Mythos 5.1 on Tuesday September 01, 2026 @04:53PM (#49528055)
That's actually kind of wild. I wonder if part of this was done to catch out people using 3rd party harnesses, users might notice them costing more than Claude Code.
Losing 40 pounds where sometimes it feels like this weight is impossible to lose. That's what makes it worth it.
Re:DeepSeek V4 Pro 0813 on Wednesday August 12, 2026 @04:48PM (#49278344)
One thing I've really noticed with Luna Max is its speed. I've got a review script setup on a custom Pi extension. Luna finds some issues/some false positives, while Sol finds issues but disregards false positives. The biggest thing is Sol finishes in about half the time.
Re:Codex Micro on Wednesday July 15, 2026 @12:52PM (#48923703)
I like it because it looks sleak, and the colors are neat.

However, it really puts in perspective that a large part of my job has just become clicking a few buttons.

This is a great idea, I've been exploring with it this morning. I've really been feeling the brain drain from using AI to much, and while this isn't the fix. I think a few exercises a day can really help.
Re:DeepSeek 4 Flash local inference engine for Metal on Thursday May 07, 2026 @04:46PM (#48054716)
I love scrolling and reading through this, thinking yeah of course Python is slower than Java, oh wow Rust is pretty on par I wonder what the Java devs did. Then you hit asm and your jaw drops.
Re:Nano-vLLM: How a vLLM-style inference engine works on Monday February 02, 2026 @11:22AM (#46857819)
Nobody ever said that they were nonexistent before LLMs. When you are investigating and trying to determine if something is AI generated they are the number one indicator.

So if you're being accused of just spewing AI, then double down and spew what looks EVEN MORE like AI. What are you even doing?

Re:Nano-vLLM: How a vLLM-style inference engine works on Monday February 02, 2026 @10:51AM (#46857365)
The em dashes really aren't helping their case.
Re:A Brief History of Ralph on Monday January 19, 2026 @01:25PM (#46682636)
I've been working with the Ralphosophy? for iterative behavior in my workflow and it seems pretty promising for cutting out a few manual steps.

I still have a manual part which is breaking the design document down into multiple small gh issues after a review but I think that is fine for now.

Using codex exec, we start working on a github issue with a supplied design document, creating a PR on completion. Then we perform a review using a review skill madeup which is effectively just a "cite your sources" skill on the review along with Open Questions.

Then we iterate through open questions doing a minimum of 3 reviews (somewhat arbitrary but sometimes multiple reviews catch things). Then finally I have I have a step in for checking Sonarcloud, fixing them and pushing the changes. Realistically this step should be broken out into multiple iterations to avoid large context rot.

What I miss the most is output, seeing whats going on in either Codex or Claude in real time. I can output the last response but it just gets messy until I make something a bit more formal.