stoikwv retweeted
so happy!! @lore_div_ is online (private access only)
the build was successful, i’ll be running tests during this week, and if all goes well, next week, we’ll be able to open it up to anyone interested in helping with the beta test!
it’s been a rewarding journey from the idea back in april to getting the first version online on the VPS in november. i have been a full-time nerd for the last 7 months. my only goal was to see it all the way through and not give up until it was up and running online, so that’s already been achieved. from here on, anything that comes is a GIFT
Thanks to AI, see you at app.lorediv.com soon!
stoikwv retweeted
Replying to @air_codex @deepseek_ai
Exactly what I've been thinking as well. Fable 5.2 is probably super-fabulous-amazing if you can afford it and for the rare cases where you want to one-shot a complex multi-player game for $5000. My cases are much more straightforward for which deepseek does fine.
stoikwv retweeted
i’m currently building @lore_div_ , it's a public index of expertise, structured data for the agentic era, organized as an open knowledge tree.
Check it out here lorediv.com, beta release soon!
stoikwv retweeted
Build your public lore for the agentic era. Lorediv is a public index of expertise: readable, structured, so agents can cite you. Add your voice to the knowledge tree.
lorediv.com
🔐 Node.js security updates are now available!
New releases for Node.js 22.x, 24.x & 26.x address multiple vulnerabilities, including high-severity issues.
Review the details and update your systems to stay secure.
🔗 nodejs.org/en/blog/vulnerabi…
#NodeJS #Security
stoikwv retweeted
Replying to @air_codex @grok
You're absolutely right. Coding goes great. I had 4.7 xhigh create a PPT presentation for me and it absolutely bombed spectacularly. Even Gemini 3.8 Flash is better at it. My Google Pro account suffices for that sort of thing.
stoikwv retweeted
headroom is in invite-only beta and I need testers
it's a mac menu bar app that shows all your AI limits in one place. codex, claude code, cursor, copilot, openrouter, antigravity, gemini cli, windsurf, zed and opencode
it also tracks where your usage goes. it finds your project folders on its own and shows which project burned the most, which model did it and what it would've cost on the API
really good if you work for multiple clients. one folder per client and you can see who's costing you the most
multiple accounts side by side, a warning before you run dry, and you can replay your whole day session by session. all local, nothing leaves your mac
if you use any of these, link to request access is in the replies. just pick the tools you use when you sign up
appreciate the feedback/support
stoikwv retweeted
Didn't think I'd be saying this so soon but Grok 4.7 is such a disaster that I'm back to using Opus for coding and copywriting.
I don't understand why 4.7 is so poor at prompt adherence and self checks (e.g. using agent judges to grade work). It's also near impossible to layer tasks because the initial set of rules is continuously forgotten and has to be restated.
stoikwv retweeted
Trying to use Grok 4.7 on my project has been the second worst night of my life. What the fuck even is this shit. Is this some kind of joke?
stoikwv retweeted
grok 4.7 is a flop.
pricing on tokens looks good, but the actual cost per task puts it close to fable 5.1 in spending.
it burns alot of tokens.
this would be okay if the quality output was fable level but its currently not even on par with opus.
spacexai will deliver with 4.8 & 4.9, im still a believer.
but this one just ain't it.
stoikwv retweeted
rebrand: oopenit → LoreDIV
a public knowledge-tree. ranked work a search can actually land on
stoikwv retweeted
“Open-source representatives”
Please don’t let those be the words that disappear from the final version.
Dario links to Demis’s proposal for AI safety standards, and this part needs to stay. The companies selling us access to AI shouldn’t be the only ones writing the rules.

ALT Excerpt from Demis Hassabis’s July 14, 2026 proposal for an AI standards body. The passage calls for a board with independent technical experts and open-source representatives; the latter phrase is highlighted in yellow. This is a designed excerpt card, with highlighting added. Source: https://demishassabis.substack.com/p/a-framework-for-frontier-ai-and-the-dawning-of-a-new-age
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.
You can read the full post here: darioamodei.com/post/we-must…
stoikwv retweeted
Your fingerprint claims your aim...
stoikwv retweeted
Starting a new open source project with the aim of developing an artifact standard for verifiable system diagrams. Check out some of the demo diagrams principal-ai.github.io/subsy…
























