Will Kurt

Ferment my own alcohol, run my own LLMs, that's just the kinda guy I am.

[Will Kurt @willkurt]

🥳Check out: Token-Explorer! 🤖
Interact with and explore LLM token generation!
Features:
- Step through token selection
- Remove tokens to explore alt paths
- Fork prompt and quickly switch between them
- Visualize all token probabilities and entropy!
- OSS (github in replies)

[kai @ssvankai]

accidentally stumbled upon this crazy paper abstract

[Andrew Ng @AndrewYNg]

The new White House policy requiring green card applicants to apply from outside the US is a capricious attack on legal immigration. It will hurt families, leave us with fewer doctors, teachers and scientists, and hurt American competitiveness in AI.

[Will Kurt @willkurt]

@mr_r0b0t @gajesh @NVIDIAAI Concurrency only matters for throughput, it's not going to impact tokens/s per request/user and you're still fundamentally limited on the size of model. Yes, the DGX at that pricepoint might be the best option, but at that bandwidth you're still very limited.

[Bryan Bischof fka Dr. Donut @BEBischof]

@hugobowne Challenged me to make a chart live with my new charting library, and here's what i managed!

[Will Kurt @willkurt]

@iamreddave Wow! That is a beautiful cover!

[Will Kurt @willkurt]

Reading two wildly different books (Žižek’s “Too late to awaken” and Smullyan’s “To mock a mockingbird”) and each makes reference to the exact same joke (with different names/attributions)!

[Will Kurt @willkurt]

This is actually an under discussed take. We make a lot of assumptions about our interiority that don’t do well under scrutiny

[Daniel Tenner @swombat]

If AI isn't conscious... maybe you're not either?

[Will Kurt @willkurt]

@oprydai How many GPUs do you have and how many requests do you expect to be serving concurrently?

If the answer to both is roughly 1 then llama.cpp is a good place to start (esp if you have < 1 GPU).

Otherwise you'll probably get more value out of vLLM

[Will Kurt @willkurt]

It's pretty funny how often on both X and LinkedIn I see posts of the nature "You CAN'T do X with Y!!!" while I am actively "Xing with Y"

[Will Kurt @willkurt]

@TheAhmadOsman Heat and electricity bills? I run all my image models through my RTX 4090, but I keep any long running LLMs running on my M3 max MBP.

Ultimately it boils down to bandwidth vs wattage

That said, I've not seen a compelling argument in favor of the DGX

[Will Kurt @willkurt]

A huge distinction in success with agents really boils down to whether or not you're solving a problem with a directly measurable outcome.

I also believe people will increasingly question why anyone is spending significant time writing software without a measurable outcome

[Will Kurt @willkurt]

Weirdly starting to enjoy X again, you just have to recognize that there are only a handful of real/interesting people left.

A few likes from the right people means what 100s from randos did a few years ago.

[Will Kurt @willkurt]

@jun_song @LottoLabs People really underestimate how much everything really boils down to memory bandwidth and power consumption.

128GB of memory isn’t much better than 24GB if your bandwidth is still ~250 GB/s and your local model isn’t really “free” if you need 1200 watts to do inference.

[Will Kurt @willkurt]

@nptacek Seriously, the last chapter of my soon-to-be released book on Stable Diffusion is all about using proprietary models to improve base SD. As long as new information can be added a model can be improved

[Will Kurt @willkurt]

This made me feel uncomfortably “seen”

[roon @tszzl]

people are walking around with their laptops slightly ajar to keep their agents running

[Bryan Bischof fka Dr. Donut @BEBischof]

When you invite @willkurt to speak to your class:

[Will Kurt @willkurt]

@jun_song So apparently cats do this because they are trying to imitate you, so if you have a old, small laptop for your cat it *might* use that instead!

[Will Kurt @willkurt]

Yes you should host your own LLM, and yes you should host your own private git server, but you should also ferment you own booze! Local apples, home fermented cider!

[Ahmad @TheAhmadOsman]

Qwen 3.6 27B means the permanent underclass thing has been canceled btw

[Will Kurt @willkurt]

@LottoLabs Exactly. I don’t need local models to be better than proprietary SotA *today*, I just need them to be as good as proprietary models where when I started to be able to reliably trust agents to do their thing.