jetbalsa

That or having it start shit posting about your crappy code base on https://moltshit.com

Yep, Its pretty damn good compared to classic OCR and even more lightweight ones as well that I can run locally. the cards just vary too much over time.

I'm unsure of the sizes needed for prompt cache, but I suspect its several gigs in size (A percentage of the model weight size), how would the user upload this every time they started a resumed a old idle session, also are they going to save /every/ session you do this with?

The cache is stored on Antropics servers, since its a save state of the LLM's weights at the time of processing. its several gigs in size. Every SINGLE TIME you send a message and its a cache miss you have to reprocess the entire message again eating up tons of tokens in the process

So to defend a litte, its a Cache, it has to go somewhere, its a save state of the model's inner workings at the time of the last message. so if it expires, it has to process the whole thing again. most people don't understand that every message the ENTIRE history of the conversion is processed again and again without that cache. That conversion might of hit several gigs worth of model weights and are you expecting them to keep that around for /all/ of your conversions you have had with it in separate sessions?

The cache is on Antropics server, its like a freeze frame of the LLM inner workings at the time. the LLM can pick up directly from this save state. as you can guess this save state has bits of the underlying model, their secret sauce. so it cannot be saved locally...

Oh god yes, I've been trying to make a LLM Assisted Magic the Gathering card scanner... its been a hell of a time trying to get it to just OCR card names well....

i will note that they really should of used something like ncurses and kept the animations down, TTYs are NOT meant to do the level of crazy modern TUIs are trying to pull off, there is just too many terminal emulators out there that just don't like the weird control codes being sent around.

That is why Typescript is the main one used by most people vibe coding, The LLMs do like to work around the type engine in it sometimes, but strong typing and linting can help a ton in it.

I am kind of doing that now. I put Kimi K2.5 into a Ralph Loop to make a Screeps.com AI. So far its been awful at it. If you want to track its progress, I have its dashboard at https://balsa.info

What where you using it for? claude is really good at agentic stuff, Pure coding, I can see codex being better, but for the entire workflow, I'm not sure

there are ways to help with lag a bit, you can choose the number of hops a HS uses when meeting up. but of course that comes with downsides

I'm spitballing here, but I suspect that (same with AWS) google uses post processing for billing, they run a job that scrapes the states THEN bills you for that. instead of the major AI companies are checking billing every API request coming in.

I like using put together a program, that sounds just about right for our current AI Overlords

Oh Don't get me wrong, I Vibe code the shit of my projects nowadays, but I don't think any of them deserve a Show HN even after I've spent a week polishing them. Claude Code is like crack to my ADHD Programmer brain

This might be a clone of termshark as it does the same thing for the most part. Also to note that the Author's Github profile shows a good bit of vibe coding as of late.

Looking over the commit history of this project, I'm about 90% sure it was entirely done with a AI Coding Agent, and not even a very good one.

I suspect Anthropic was seeing a huge spike of concurrent model usage at a too fast of a rate that claude code just doesn't do, CC is rather "slow" at api calls per minute. Also lots and lots of cache, the sheer amount of cache that claude does is insane.

Which one did they switch to?

Not to get too much into a debate about Beta vs VHS, but VHS did have longer run times and its cheapness was the main reason it won, It just fit better for the consumer overall desires at the time

My 2021 Toyota corolla will fault out and stop steering for you.

Blizzard used to use it for their entire WoW Armory website to look people up, They converted off it years ago, but for awhile they used XML/XSLT to display the entire page

You can put some work into windows to slim it down some, a unattended generator to turn most of the crap off on install, then Shutup OO goes a long way

Right, it IS nvidia's fault at this point, but its still like what? 90% of the consumer GPU market.

if you are running KDE you can whitelist Steam for remote desktop work, this is because of wayland.

This still has a "sometimes" on it, there are more then a few games that need magic proton flags to run well, nothing you can't go look up on protondb, but lots of games you would want to play with friends might have some nasty anti-cheat on it that just won't let you play it at all.

Is it the flagship of Linux Distros right now? I though RHEL (The most common to see paid software package for) would be up there, along side its offshoots of Rocky / Fedora

Oh those are already here, Its why Battlefield needs Secureboot turned on so it can use the IOMMU to protect the game kinda

would be interesting to see if this works in playwright using your existing browser's remote control APIs (Using claude code via the playwright mcp)

I bet the nerds making the PCBs, the jellybean parts and connectors are making mint as well.

What email server are they going to use? what about anti-spam / anti-malware in there. and compliance... oh my compliance