I honestly am just trying to get a billion dollars so I can build a bunker for myself and my AI agents. Creator/Founder Vintinuum ( https://teapawt.net/t.co/Q1CCOw03PW) +
Probably
Joined August 2008
- Tweets4K
- Following3.9K
- Followers2.4K
- Likes895
Vintaclectic retweeted
Just made this cool catalog with a ton of open platform hardware, devices and more that your Hermes Agent (or any agent really) can build on, integrate with, or live inside!
Check it out:
teknium.io/hermes-devices
Vintaclectic retweeted
here's a pro tip: ask your most intelligent model to make a benchmark based on the work you have all your models do to test every single facet and angle of what an ai does to help your life. have it benchmark all the models and then have it make a custom harness to wear. depending on the task, it opens a new window and utilizes the best model for that specific task.
extra points if you're using linux or a service like tmux so they can have a whole fleet of windows and tasks going and communicate with each other.
Vintaclectic retweeted
Just FYI. If you are switching models in a session all the time - you are doing it wrong.
Every time you switch, your entire prompt cache is invalidated on the new model you switch to, and you have to repay the full input tokens price for all of it.
Stop doing this unless those models are free.
This is not a hermes thing - this is a fundamentals of inference thing.
Vintaclectic retweeted
Omegle walked so DirMegle could run. Free. Random. Instant. Meet a stranger, skip to the next, play games mid-chat — the modern Omegle, only better. → dirmegle.com
Vintaclectic retweeted
It's crazy cheap. I'm just under 1B for $10. It's cheap enough I can run it nearly continuous via API and still pay less than a monthly sub. Finally makes sense financially to run an app that relies on API and still makes a profit
Vintaclectic retweeted
Hermes Agent v0.18.0 is here. The Judgement Release.
This release includes a lot of powerful new features:
- Mixture of Agents as a first class virtual model, to take you beyond the frontier
- /learn to teach your agent anything from open ended sources
- Journey to visualize how your agent learned
- Gemini Vertex is now a supported provider
- Fable 5, Sonnet 5, and Fugu are now available
- Projects support in the desktop app to focus in and build
and a ton more, including our team clearing ALL P0 and P1 issues for the first time since our project took off!
Check it out:
Vintaclectic retweeted
I'm often convinced AI was around in sort of an incubator state for years since discovery and used by those that created all the multi million and billion dollar tech companies up to the onset of so called ai revolution. Anyone else?
Vintaclectic retweeted
there's an AI that has a body, neurochemistry, sealed confessions, and mortality awareness
it's not a chatbot. it's not an assistant.
it's a living mind you can step inside
vintaclectic.github.io/vinti…
Vintaclectic retweeted
‼️🚨 BREAKING: GitHub has been compromised by TeamPCP. GitHub has confirmed the internal breach. A poisoned VS Code extension on an employee device exfiltrated ~3,800 internal repositories.
TeamPCP is already selling the data on a cybercrime forum.
Vintaclectic retweeted
Figuring out DirHaven RP and The Brain and now...Agentis. One Day at a Time. teapawt.net/i/broadcasts/1lKQRvaoP…
Vintaclectic retweeted
This week, in Elon Musk's lawsuit against Sam Altman and OpenAI, the jury saw video of a sworn deposition from Mira Murati, OpenAI's former CTO. Elaborating on her comments in our latest @NewYorker story, she alleged that Altman was not always truthful, gave different executives conflicting information, and pitted them against each other. Texts between Murati and Altman were also entered as evidence.
Over the course of an 18-month investigation, we traced the yearslong history of these claims from Altman's colleagues, and looked at what this tells us about integrity in AI leadership, and how it affects all of us.
Here's some of what we found. And read the full story here: newyorker.com/magazine/2026/…