481–490 of 965 posts
You techbros need to get off your ass and go to work.
Earlier quoted context omitted.
I've traveled to China ~5x [0], visited a range of cities Tier 1-3 over a collective 5 months, and grew up in the USA. I also currently live in Vietnam (~3.5 years) and spent 5.5 years working for a Singaporean company and a team stationed in Beijing. I don't really know how else to express my experiences living in, working with, and interacting with people in both of these countries. Perhaps you can share how life w…
I am not really interested in reality like this. I just repeat things I read on my social media feed. America bad. China good.
I am afraid — if Chinese models go mainstream it has a clear way of pushing its narrative way beyond its otherwise borders. More like a Trojan horse it is for the Chinese. Here is a quick example of how Chinese deepseeks agent works kn its underlying model) when asked a tough question https://x.com/jinen83/status/2079406993979383902?s=46&t=D7hQ...
Same thing happen for western models, try to ask about Gaza genocide and see for what side it will stand
We have got very far from Cicero's coining of the word 'intelligentia' (from inter legere, a 'reading between' and hence discernment) when people talk about 'intelligence' as a commodity
People have been decrying the 'cheapening' of the word intelligence for over a century now, going back to Psychology's adoption of the word and coining of nonsenses like "Intelligence Quotient". "Artificial Intelligence" is just the latest degradation of the original humanistic meaning, and now people aren't ever bothering to prepend 'artificial' to their idiotic use of the word
Earlier quoted context omitted.
Correct. We need open weights, open code and open data. If nobody else can reproduce what someone did there will always be security questions. Even if we can reproduce it there could still be security concerns but it's more realistic to investigate yourself.
Exactly what are the possible 'security issues' of self hosting an open weights model?
* Me, as an individual, because I might not be able to pay price hikes, because my revenue (salary) is much lower than what they want and I can't support my expenses via huge bank loans.
* Again, me as a new entrant to the industry, LLMs are basically pay-to-play games, again related to price hikes, new entrants might not be able to afford paying those prices 24/7 - which you need when learning new things.
* Any non-US company, US can block the models which can disrupt the whole business.
* Even some US companies, for example if you operate in EU and EU somewhat changes their mind and follow the ICC and require you to stop working with Netanyahu (war criminal as per ICC), then following laws in EU, might create trouble to your whole business.
Earlier quoted context omitted.
Their kv cache is smaller so they can use less vram and also keep your prefix cached for longer. https://deepseek.ai/blog/deepseek-v4-compressed-attention
This also comes with significant capability reduction. deepseek-v4-flash is very good in the [edit] This is my observation from using it without an specific context engineering to optimize for Deepseek's cache compression and sparse attention mechanisms. I am pretty sure that if you specifically structure your context to align to the cache compression boundaries you can significantly improve performance in the full 1…
Switch to a modern sampler like min_p or ideally a better one like top-n-sigma (it’s in llamacpp) and your “my model gets stupid at long context problems” will basically go away.
Unfortunately this fact is still not well appreciated yet despite nearly every modern sampling technique getting an oral wherever they get presented. Min-K just got an oral at ACL 2026, for a hyper recent example of this. There’s a reason they keep getting orals.
The field massively ignored sampling for mostly safety reasons and now the whole field incorrectly believes long context doesn’t work on small models. Long context is an out-of-distribution problem. Your sampler configured properly keeps you in distribution.
Oh and this is doubly true for quantized models. I run my qwen 3.6 27b with 4bit quants from unsloth and get excellent performance because my sampler stack is good and not the garbage that is top_p and top_k. Also, yes, you need to ignore the trash recommended sampler settings from the Chinese labs (they’re wrong/bad).
The 2 things people need to remember: 1) China can (and does) use the models to influence the west. They train in false information about Taiwan and Hong Kong. Or pretend like history is in favor of China. 2) Ignoring the models containing false information, they are incredible. But you should be scared of running inference via the model creators directly. If you think your data is safe compared to running it via mod…
Regarding point 2, I don't trust my data being safe running inference on model creators api, but neither do I trust US providers. Both use it for their own benefit, the only difference is the country of origin. The US has a lot more legal safeguards for this but I don't trust they don't do it regardless.
Earlier quoted context omitted.
It’s not hypothetical. Magic strings are a known and implemented feature for standard model interaction. Nearly impossible to detect unless you know where to look with current technology.
Maybe I should clarify. As I understand it, the kind of vulnerability being discussed is something like a Chinese model invisibly "realizing" that it's working on an American project, and then deliberately leaving subtle security bugs in its generated code for Chinese hackers to later exploit. As far as I know, that scenario is hypothetically possible, but has never been demonstrated to happen in the wild. Admittedly…
Earlier quoted context omitted.
The distillation explanation is classic American exceptionalism: No one could possibly do anything unless they were copying American leaders (where "American" means a bunch of Chinese, Canadian, Europeans and Indians working in the US). It's also a bit of securities defensiveness. Pretending that you really do have a super moat, people just keep swimming in it so you just need to add more alligators. It's farcical. A…
> Previously Anthropic has reported on some Chinese firms doing chicken-shit level of API calls, that at most would be doing some Q and A or final fine tuning "Anthropic said the campaign was conducted between April 22 and June 5, 2026, and generated more than 28.8 million exchanges with Claude through almost 25,000 fraudulent accounts." I don't know why you're trying to downplay it. European models are so far behind…
Ignoring that I have literally zero trust in anything Anthropic has to say on this -- they have been doing the hysterical routine and trying to get every bit of government granted monopoly they can[1] -- those numbers still simply aren't that impressive.
>European models are so far behind because...
What a non-sequitur. Europe, like much of the West, foolishly delegated tech, media, payment systems, etc, to the United States. European efforts on this are poorly funded, poorly capitalized, and marginal efforts.
China is very much not Europe. China is looking to leave the US to the dustbin of history, and their efforts are a little more concerted.
[1] Surely Americans are aware that Anthropic and OpenAI are both very close to getting the US government to ban and fully criminalize the open Chinese models, right?