You can run the quantized versions of DeepSeek locally with normal hardware just fine, even with very good performance. I have it running just now. With a decent consumer gaming GPU you can already get quite far.
It is quite interesting that this censorship survives quantization, perhaps the larger versions censor even more. But yes, there probably is an extra step that detects "controversial content" and then overwrites the output.
Since the data feeding DeepSeek is public, you can correct the censorship by building your own model. For that you need considerably more compute power though. Still, for the "small man", what they released is quite helpful despite the censorship.
At least you can retrace how it ends up in the model, which isn't true for most other open weight models, that cannot release their training data due to numerous reasons beyond "they don't want to".
Mastodon mods being that happy to ban people is the reason for me I never even bothered. And their behavior reflects on any instance, technically correct or not.
I can just as well use a Discord channel because economic interests are at least predictable and the same rules apply to anyone more or less. Or I could use a reddit sub with my political alignment because mods there are equally ban happy as well.
I would take a lot to convince me that members on prominent Mastodon instances are curious about other opinions, but something in the larger picture just doesn't add up.
That is fine though, but Mastodon currently cannot be a place for everyone, regardless of technical possibilities. And it should be said that a lot of Mastodon users were involved in the witch hunt against Stallman, so at least some of the prominent users seem to be toxic.
The CEO did gave a statement about their motivation. Could be a lie, but he delivered and it is also vastly more sensible that what we often hear from other companies. Google and Meta are an exception for this space though.
Also, because not only the weights, but also the data is open, any propaganda can be identified and corrected. This is not the case for other models and what we have seen from Gemini, there certainly are "adaptations". I don't think Google had ill intent here, but this would fit what some would classify as propaganda.
Which is ironic because Google needs to improve their reputation about sunsetting early. This is one of the main arguments for why many businesses for why they do not employ their alternatives
The thinking is quite fascinating though, I love reading it. Especially when it notices something must be wrong. It will probably be very helpful to refine answer for itself and other models.
It does add latency of course, but I still think that I could provide all AI needs of my company (industrial production) with a simple older off the shelf PC. My GPU is decently recent, but the smallest model of the series and otherwise the machine is a rusty bucket.
I didn't test it thoroughly yet, but I have some invoices where I need to extract info and it did a perfect job until now. But I don't think there is any LLM yet that can do that without someone checking the output.
That was perhaps a bit too general, but aside from meta and Google they didn't share their research and tried to sell AI products as fast as possible and tried to lobby legislation to keep their head start. I would also include nvidia here, that has some moat through software integrations.
I haven't tested deepseek for censorship yet, but they shared their release and even their input data. And in this case you could correct its shortcomings, so propaganda would be difficult.
Not on my machine currently, I use the 14b Q4 model I think, which delivers very good answers. I run a 4060 with 16gb memory and performance is quite good. I used the largest model that was recommended with this amount of VRAM, I think it was the 14b one.
I do have some applications that process images, text and pdf files and I use smaller models for extracting embeddings. I think my system wouldn't be able to handle it with decent speed otherwise.
I do run LLM on a M1 16gb macbook air and performance is surprisingly good. Not for image synthesis though and a PC with a dedicated GPU is still significantly faster with LLM responses as well. Haven't tried to run deepseek on the macbook yet.
Oh yes, I am firmly on Team China here because US companies got too greedy. Meta is an exception here though and they also propelled AI development massively.
DeepSeek is awesome. Any AI task yet implemented in our business can be run from my local PC with just the smaller models. And my PC is fairly crappy to begin with.
OpenAI looks quite silly with their "we have to close everything".
Maybe in EU legislation where this privileged access is now mandated by law for every certified fact checker.
Although these aren't conventional fact checkers, they just "flag" content.
> Trusted flaggers are public and private actors with privileged access to internet platforms through which they can identify illegal and harmful content. Before DSA, trusted flaggers existed as a voluntary system.
They are not independent enough and would develop their own interests.
> narrative of persecution that serves fascistic rhetorical goals.
Yeah, fascists are known to flag content too. Otherwise this is complete conjecture.
> no institution can be trusted
No institution, which will always be dependent on government funding and special legislation, can be trusted to moderate internet content. That much is true.
To be honest, they aren't the ones that started it. I remember sites like reddit removing subs with fans of Donald Trump and Twitter censorships was very prevalent as well. You might disagree that liking Trump is a good choice, I cannot dispute that, but nevertheless people fell over themselves to ban it everywhere.
I don't think this kind of revanchist behavior is sensible, but it also isn't "blocking democracy" either. Or at least someone would need an honest attempt explaining why such removals are unjustified and in a way that addresses shortcomings here for all platforms and political directions.
Government influence on online discourse is a negative, you could argue that voting for an idiot is then the next best compromise.
I believe there was a lot of interest that wanted to contain user voices as much as possible.
Some say there would be something like a professional critic. For literature that certainly is true since you can only draw comparisons by being familiar with a broad selection of media.
But on the other hand, one that is sadly more relevant these days, such critics aren't necessarily independent and can more easily be manipulated by the usual media power players or other interests.
Surprisingly IMDB still was very useful at first, despite being bought by Amazon pretty early. But they more and more removed user content and now their reviews are as authentic as biting on a piece of plastic.
IMDB got less relevant, rotten tomatoes got more popular. Now the mistakes are repeated and we hopefully get a more free platform again soon. I really like the old style "every user has equal weight" approach. I accept that there is brigading and spam attacks that can manipulate a score. That is information as well and it is easy to see the patterns and I probably watch something that is a polarizing 5/10.
I think they have a responsibility to honor the constitutions of respective countries and not do that at all.
Seriously, this undermines my belief that an international institution like Europol can work if not heavily constrained to be a bridge between executive forces of cooperating countries.
Some members of Europol do good work to fight international crime, which includes criminal online networks. I believe they need wiser leadership that accepts constraints and realities of their tools and use these to the best of their abilities.
Encryption, including end-to-end encryption for confidential communication is a very important mechanism for democracies that cannot be given away for ambitions of crime fighting organizations that do not have democratic legitimacy to access any form of communication.
I doubt this is meant as a nazi salute at all. This is click and outrage bait. I did not flag this article, but I believe many question the honesty here.
Sure, you could say that only running the 600+b model is running "the real thing"...