Hacker Newsnew | past | comments | ask | show | jobs | submit | indigodaddy's commentslogin

This is super cool.

Not affiliated but trymyrepo.com is very neat as well. Seems similar space.

Also I had Muse agent create a skill that can do videos. Worked pretty well. Things are getting wild.

Lastly, this might fun to those who like this space: (my thing)

https://github.com/jgbrwn/mst3k-anything


Check out Neuralwatt. They are ZDR and great energy based K3 pricing. They also have a K3-fast which is basically no reasoning (in addition to regular K3).

that is nothing like how we consume Ai

industry standard is price-per-1M tokens, don't do something different, even Google caved and moved from their char based pricing to tokens (the fundamental unit of computation in ai)

GPUs are rented in $/h, like every other piece of hardware in cloud


Why are you assuming the gpus are rented?

Anyway, they still have token based pricing if you prefer. The energy pricing is often cheaper though


Not interested, it looks amateur

A trust page is required like https://trust.fireworks.ai

Very happy with my $10/month OpenCode Go sub for personal use

https://trust.opencode.ai/


Energy based is way more fundamental than token based. To each their own.

Wonder how this will affect my free tier VM with 768M ram that is usually about $3.5x a month. Not sure how much of the price slice on that that memory takes up, but even adding $.70 I'm still well under $5. Hope my mental math works out or I guess it's time to try Python Workers.. :)


No FastHTML?


Jeezus h the ads ruin it W H Y

Is it really worth the potential of like $1 month if you're rockin the shit out of it, when you can host for free or basically free anywhere these days

Oh I checked whois on one of the IPs and they are using square space. I guess that's why. Gotta recoup something from inexplicably deciding to use squarespace


Not using an adblocker in 2026 AD is just asking for it


This is gold, tysm for posting


His whole channel is full of commentary like this. It's simultaneously funny and really informative about the sort of blend of grift and AI psychosis that is genuinely out there.


oAI isn't anywhere near close to fucked as long as their models are head and shoulders above even the very best open models in terms of tool calling and rock solid stability/reliability for agents/coding harnesses. Which, they are right now and we'll see if open models actually catch up in that regard. Even the "best" open models pale in comparison with tool calling and general "prompt and go do something else for an hour" reliability that we have with GPT models. With GPT models, streaming rarely stops unexpectedly. You almost never have to constantly nudge them along, etc. Granted with open models all of this can vary depending on the provider, and perhaps open models/protocols/APIs/harnesses aren't well enough aligned, but OpenAI models just seem to work without constant (or hardly any) wrinkles and with almost any harness/agent.


>oAI isn't anywhere near close to fucked as long as their models are head and shoulders above even the very best open models in terms of tool calling and rock solid stability/reliability for agents/coding harnesses

That's already not the case today. If you sat me in front of an LLM and told me to figure out if I'm working with K3 or Astra, I could probably do it, but it would take some work to be certain.


Well I could for sure. I guess a lot of this is indeed very anecdotal.


All we do with these things is some work though


404


I used to use x2goserver which was and still is awesome.

These days I prefer docker-based KASM Workspaces, or if you just want a nice GUI over your "bare-metal" VM or server then KasmVNC is great and easy to setup (not that KASM Workspaces is hard to setup)

https://hub.docker.com/r/linuxserver/kasm https://github.com/kasmtech/KasmVNC


So, I think one of the more interesting fits for AI is for tech-heads that aren't really proper devs/programmers. So like for example myself I'm just a Linux/sysadmin/devops guy. I can figure some shit out for sure and have glided upon the surface of webdev for a few decades, but I cannot by any means call myself a developer. However, with AI over the last couple of years, I've been able to do a lot of cool shit that I never would have dreamed of, and at the same time I've learnt some things (I won't claim to think that I've learned more than I could/would have if I'd actually put in the study/time/work, but that might be a different conversation). Point is for non-dev tech-heads like myself, I think that might be the sweetest spot for the actual enjoyment of AI? Eg we can perhaps put out some slightly above average AI slop, yeah it's not exactly "ours," but goddam did we have fun and we did some shit that we've always wanted to do, and it came out pretty damn decent, good enough for our usage!

Some things I've done that have been incredibly fun and invigorating and never would have happened if not for AI:

https://afterword.exe.xyz (I wanted badly to play Jonah's original Word After Word (desktop only) game on my phone or just in a web browser and with remote friends/family, so got his blessing for a web remake)

https://github.com/jgbrwn/fetchcade

https://loderunner2099.exe.xyz/

https://v2.my-upc.com (don't ask, if you know, you know)

https://cartalk-amp.exe.xyz

https://aiconversation.apps.techpique.com

https://github.com/jgbrwn/mst3k-anything (Paste a video URL. Get back a robot-heckled version)

https://hn-readtheroom.exe.xyz/

https://jgbrwn.itch.io/cannonsmash-web

https://history-of-webhosting.exe.xyz/

https://wordizzle.exe.xyz/ (WIP)

https://exe-muttha-fukken-dev.exe.xyz/ (fansite lol)

https://github.com/jgbrwn/vibebin

https://github.com/jgbrwn/rsync-web

https://github.com/jgbrwn/retro-computing-stream

https://github.com/jgbrwn/github-repo-stream


Consider applying for YC's Winter 2027 batch! Applications are open till November 2.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: