Hacker News: Tenoke

New comment by Tenoke in "Mistral raises €3B"

Tenoke — Tue, 08 Sep 2026 10:00:35 +0000

It's possible to be a waste of European money that could be better used elsewhere, though. I think the same about LeCun's company sucking up the little funding here.

New comment by Tenoke in "Mistral raises €3B"

Tenoke — Tue, 08 Sep 2026 09:59:21 +0000

> Europe absolutely needs a home-grown AI lab, especially with Pax Americana looking increasingly shaky.

The only problem (at least in the LLM space) is that you can do more in Europe by just getting the best Chinese Open weights model (do a finetune if you really want) and do more for cheaper than using Mistral.

New comment by Tenoke in "Any Human Ever – One life, drawn at random from all who have ever lived"

Tenoke — Thu, 03 Sep 2026 18:03:40 +0000

Yes, this is exactly what I expected (showing the anthropic principle in action), so I am disappointed they are not sampling correctly.

New comment by Tenoke in "Samsung's Processing-in-Memory (PIM)"

Tenoke — Sat, 29 Aug 2026 11:52:37 +0000

>there ought to be a tipping point beyond which local inference is good enough

There's no such ought really. Even at current levels you'd need like a 100x gain from here to approach current top proprietary models (probably a lot more for say Mythos or Mythos 2), and it's not like they are stoppng to improve. This is before we even account that you'd just be running 1 agent then, and not a swarm like you'd be able to in the cloud or that you can do only so much compression before you are losing out

New comment by Tenoke in "Samsung's Processing-in-Memory (PIM)"

Tenoke — Sat, 29 Aug 2026 11:04:00 +0000

There's been a ton of optimizations already, it hasn't remotely reduced demand even temporarily. More efficiency just makes the compute have even higher ROI per $ and watt spent.

New comment by Tenoke in "Nvidia agrees to acquire Hugging Face for $13B"

Tenoke — Thu, 27 Aug 2026 12:32:03 +0000

Too base cynicism. I'd be willing to bet you my $200 to your $100 that doesn't happen.

New comment by Tenoke in "Don't Paste the AI, please"

Tenoke — Thu, 20 Aug 2026 12:23:41 +0000

That's a good formalization of it. I will think you are a bit of a fool if you dont use AI but you guarantee you'll be a fool if you only use AI with no value added by yourself.

New comment by Tenoke in "Claude Opus 5"

Tenoke — Fri, 24 Jul 2026 18:05:21 +0000

Fable has more parameters. In practice it's not yet clear which one would be better for different usecases yet but they are more different than one being strictly better.

New comment by Tenoke in "Flux 3 X Mimic: The Next Generation of Video-Action Models"

Tenoke — Fri, 24 Jul 2026 12:53:01 +0000

Mistral is much worse in its respective field than Flux in their own so I hope not.

New comment by Tenoke in "Flux 3"

Tenoke — Fri, 24 Jul 2026 08:25:28 +0000

Flux 2 Dev Klein has practically been the best you could use on most commercial hardware so I really hope Flux 3 has a comparable updated open-weights model to it. if not it'd be a great loss to most hobbyists.

New comment by Tenoke in "Startup founders urge U.S. government not to shut off Chinese open weight AI"

Tenoke — Thu, 23 Jul 2026 15:57:01 +0000

Y Combinator the company doesnt particularly have to share the opinions of hackernews the public site.

New comment by Tenoke in "OpenAI and Anthropic unite against open-weight AI risks to their bottom line"

Tenoke — Thu, 23 Jul 2026 15:46:05 +0000

Because sadly, as much as I wish it wasn't the case, offense is easier, more impactful than defense.

New comment by Tenoke in "OpenAI and Anthropic unite against open-weight AI risks to their bottom line"

Tenoke — Thu, 23 Jul 2026 14:02:06 +0000

I am very pro open source models - I use them every single day.. But we obviously don't want everyone to have capabilities like and beyond what caused the huggingface incident in every domain, so it's not like it all comes from bottom line cynicism.

New comment by Tenoke in "OpenAI and Hugging Face address security incident during model evaluation"

Tenoke — Wed, 22 Jul 2026 08:33:40 +0000

There's ways to make sure env vars get only injected at runtime and arent easily accessible otherwise or to even make them inaccessible to the user your agent is running on, and for you to manually run the code with the right permissions when the keys actually need to be used. Almost nobody bothers doing it though.

New comment by Tenoke in "OpenAI and Hugging Face address security incident during model evaluation"

Tenoke — Tue, 21 Jul 2026 21:14:32 +0000

That's kind of insane. Natural that it's happened, sure, but insane. I know people don't like thinking of it like that, but things analogous to this can easily happen in various domains with today/tomorrow's models given access and a different task.

New comment by Tenoke in "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"

Tenoke — Tue, 21 Jul 2026 16:15:16 +0000

It's also very possible that they know their big model underperforms chatgpt 5.6 and fable by too much, so they are focusing on what they can get wins in like speed instead.

New comment by Tenoke in "Fable 5 vs. GPT-5.6 Sol on an NP-Hard Problem: Does /goal help?"

Tenoke — Sun, 19 Jul 2026 11:26:17 +0000

I do that but it ends up putting so much extra crap there that it has the same drawbacks.

New comment by Tenoke in "Fable 5 vs. GPT-5.6 Sol on an NP-Hard Problem: Does /goal help?"

Tenoke — Sat, 18 Jul 2026 13:10:20 +0000

Claude seems to forget what you tell it in very long work sessions (things that take weeks to develop), no matter how many times you tell it which part is extra important. I dont use goal (I guess I should), but presumably it makes it actually remember the most important instruction. I believe this here is about shorter sessions where the issue doesn't crop up as much.

Apple M7 Ultra Chip Planned with Up to 1.5 TB of Unified Memory

Tenoke — Mon, 13 Jul 2026 14:20:15 +0000

Article URL: https://www.techpowerup.com/350711/apple-m7-ultra-chip-planned-with-up-to-1-5-tb-of-unified-memory

Comments URL: https://news.ycombinator.com/item?id=48893163

Points: 5

# Comments: 2

New comment by Tenoke in "GPT-5.6"

Tenoke — Fri, 10 Jul 2026 05:03:41 +0000

>The human brain manages to self-organize with only a fraction of the information that LLMs get trained on.

So? The question isnt can we get to ASI as efficiently as a brain, the question is can we get there, which we likely can. The inefficiencies can also be fixed after that.

>No matter how many trillions of dollars get thrown at the problem, they still don't learn like humans do.

Again, so? Humans are efficient but also bad at many things that transformers are already better at because of it. You are looking at the wrong thing if you think it needs to be like humans.