Self-sovereign NixOS VM stack for agentic coding and privacy tasks.
self-hosted
Joined July 2026
- Tweets244
- Following43
- Followers18
- Likes105
Allod retweeted
Run GLM 5.3 Flash EXL3 with TensorFold ⚡️
This is a completely new recipe that ushers a whole new level of performance for @NVIDIAAI 2x DGX Sparks.
Conservative default for stability:
- 1M context by default
- 2.7M KV cache pool (!)
- Yes, it's not a typo - 2.7M KV in just two Sparks
- 4 concurrent streams by default
Performance:
- 60 tok/s on prose, single stream.
- 108 tok/s on prose, 4 concurrent streams.
~1,950 prefill tok/s for most context lengths.
Stress-tested to handle a variety of workflows.
This is by far the BEST model to run if you have two DGX Sparks. Extremely smooth experience!
Expect further improvements!
Thanks to @ashxhart for developing such a powerful engine! TensorFold will be used in many of my upcoming recipes.
Get it here:
github.com/MiaAI-Lab/GLM-5.3…
Allod retweeted
If you are interested in getting started with Local AI and have a budget between 5k-10k I recommend the sparks
Amazing building blocks and Mia’s repos + Brandon/turboderp/David’s quants are enough to have a smooth experience for 50$ a month of power
Allod retweeted
I talked with my Uber driver about AI while on the way to the airport in Minneapolis. He thought AI was going to kill us all. I told him that I thought otherwise. I said:
The thing to fear the most is a handful of companies and governments controlling the most capable intelligence on Earth. We've seen what they do with power.
Open source and open weight models are how we stay free from oppression. It's how we empower all of us.
Allod retweeted
People underappreciate the privacy aspect of local AI models and mostly talk about price.
There are things I don't ever want to expose to any cloud API. Ever. No matter its privacy policy says.
Run models locally if you care about your sensitive data, there is no other option!
The narrative of "is it worth spending X amount of $ to run local models" should change to "is it important enough that my data stays for my eyes only?"
Allod retweeted
Been noticing this more and more:
AI gives talented people insane leverage when they work for themselves.
Put them inside a big company, and somehow they’re still stuck moving at company speed.
That’s why the best talent isn’t just switching companies anymore.
They’re leaving them.
Allod retweeted
because open models don't refuse to do things for you and the government can't stop you
Allod retweeted
‼️ BREAKING: An active npm supply chain attack has compromised at least 868 packages carrying over 2 billion monthly installs with a credential-stealing worm. Shai-Hulud is back.
It started with the compromise of the GitHub account of the maintainer behind keyv, a library with roughly 127 million weekly npm downloads.
A preinstall hook fires on npm install and drops a stealer that sweeps npm, GitHub, AWS, Kubernetes and Vault secrets, and then spreads to more maintainers.
