Ozma - rock and roll pt. 3, remains one of the best rock albums anyone should run to listen to if you...
1. Don't know what I'm talking about
2. Do know what I'm talking about
🤟
@Teknium Appreciate this. Might give this another look once it's out. Cli remains primary, but I like toys. Lul. Outgoing desktop felt overweight. Cool portal, just a lot going on. Looking forward to the next.
Having sex as a husband in the 90s:
Husband: Kids are asleep. Come here.
Wife: Give me two minutes, let me turn off the lights.
Having sex as a husband in 2026:
BREAKING: Claude Code found 14,219 photos of my face on my Android that I never took. Phone was secretly taking them almost every time I picked it up. Even when I was just lying in bed. Even at home.
Not screenshots. Real frames from the front camera. On the train. In the car. At home. When I just grabbed it to check the time.
They were just sitting in a system cache. Timestamps and GPS on them. Nothing in Gallery. Nothing in Recently Deleted.
Plugged the phone into my laptop, made a full backup, told Claude Code to look for any front-camera files that weren’t in my photos. It pulled /data/system/users/0/faceunlock/cache and another folder inside Google Play Services. Over 14 thousand files. Normal ones and infrared from Face Unlock. Every single one had time and location.
Started scrolling. Almost everything from the last 15 months. Then I saw a file created 2 seconds ago. I was still looking at the folder. Phone kept taking photos while I was reading this.
Google and the phone makers say Face Unlock turns everything into a mathematical model and deletes the image right away. On this phone the raw frames were just sitting there. Some of the processes writing to that folder were from system diagnostics. Couple of old apps that still had camera permission could also get in.
Turned off diagnostics, revoked camera access from everything that doesn’t need it, cleared the cache. But how many of these frames already went somewhere else, I don’t know.
If you want to check yours, plug it in, make a full backup or use ADB, and ask Claude to look for front-camera files outside your Gallery. Path above is where mine were.
❤️
I'm pulling so fucking hard for what is quickly becoming this era's Linux. This old dog has a soft spot for community driven open-ness. Love to see it.
Jensen closed his GTC Berlin 2026 keynote with this Hermes Agent demo as the finale.
He introduced it by saying the way to build an agent is to ask an agent to build it for you, so he installed and set up a customized Hermes in one prompt.
Then he used Hermes, running on @nvidia RTX Spark, to create a simulated training environment for the @huggingface Microduck. Hermes planned an obstacle course with Nemotron and had GPT-6 Astra build it using FreeCAD, Omniverse libraries, Blender, Newton, and Isaac Lab. Hermes trained thousands of simulated ducks on that course until the validated policy was ready to run on the real robot in the physical world.
At Nous, we believe open source is a practical tool as much as a philosophical ideal. It gives users the ability to build powerful, complex workflows chaining numerous local+cloud models, devices, and tools themselves instead of waiting for a company to enable them.
It's been lovely watching hermes break out the way it has. Imho, the way forward has always been model-agnostic agent harnessing. Not for nothing, but there's a reason I remain f**king with nous well after serving 20+ models on my distributed local.
Watching pm openings 🥸
Hermes Agent is a launch partner for @nvidia's RTX Spark across the new generation of laptops.
Pick it in the built-in @Windows agent setup flow, run the new generative onboarding, and get straight to work locally or in the cloud.
Linus is impressed!
Can confirm. I tried the desktop offering and it wasn't for me. Considering that I've been building my own harness for my specific setup + workflow... With CLI, it's important to highlight that one sentence. Hermes CLI will remain as such, regardless of my build bc of its intent
We aren't giving up on consumers. But they are dumb and I never made hermes for that purpose. I made hermes agent as a coding harness for me. That's why the CLI came out first, and then remote gateway channels, and then, our first actual move toward consumer with the desktop app.
If anything, the mobile app means we are going way harder on consumers.
I am not leading the mobile app, and I dont even lead the desktop app.
My focus is and always has been on making hermes the **best** agent. Not an agent ux for normies.
MoA is not for normies. 1000's of subagents is not for normies. The intense work we've done to improve terminal and coding capabilities of hermes is not for normies.
It has always been this way, and anyone who thought otherwise was mistaken.
Getting the dgx spark block to a stable 77 tok/s at 8 concurrent, leaves plenty of room for style cache tuning and the achievement of a 600k ceiling.
Local. Sovereign. 🏴☠️🦜
Always. Although I have a fairly pronounced bias against a lot of creative direction these days. Hell, if I could adjust the terminal output shading I would. I get why companies choose more natural orange and yellow hues... But I'd love more cool tones and greys.
Software engineers are supposed to do exactly what we’ve always done. Ensure that the composition and structure of the system is robust and maintainable and that the behavior of the system properly serves the needs of our users.
This requires engineering discipline. It need not require detailed reading of the code; just as using compilers does not require detailed reading of the binary.
Small bird, fast wings, Kolibri is here.
78B parameters. 3.46B active. Up to 1M tokens of context. Built in Europe.
Now the weights are yours. Run it on your own hardware, under Apache 2.0.
@alexalbert__ I love this class of image -> GTA ieas. Imagine taking arbitrary images/videos (e.g this famous historical one from 1906 SF) and use them to build full-fledged custom GTAs, then drop into the world, move around, etc.
Are there risks to AI? Of course there are. Every tool has risks. Power tools have bigger risks. So of course we should implement guard rails. Do those guard rails mean that we slow down and stop. No! We go as hard and as fast as we can. We acknowledge the risk and mitigate it to the best of our ability. And we keep on pushing. We are after all humans. That’s what humans do.
Imho, the same as if I had factors of magnitude more. Says a lot about a person's character if they would have to shift anything because of fake Internet points.