Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

DeepSeek v4

api-docs.deepseek.com

Post reply on HN
Print view View original

711–720 of 1001 posts

Re: DeepSeek v4

#711

Post by WarmWash »

revolvingthrow wrote: post #160

> pricing "Pro" $3.48 / 1M output tokens vs $4.40 I’d like somebody to explain to me how the endless comments of "bleeding edge labs are subsidizing the inference at an insane rate" make sense in light of a humongous model like v4 pro being $4 per 1M. I’d bet even the subscriptions are profitable, much less the API prices. edit: $1.74/M input $3.48/M output on OpenRouter

Because you are comparing China to the US.

In China you need to appease state goals. In the US you need to appease investor goals.

China will keep funding them regardless of their income, because the goal is (ostensibly) a state AGI/ASI. In the US, the goal is an ROI which may or may not come with AGI/ASI.

They are different economies with different goals. We can look at past Chinese national projects and see that they are fine with burning $50 to get [social goal] that's worth $5.

Re: DeepSeek v4

#712

Post by khalic »

SquareWheel wrote: post #499
khalic wrote: post #473

Earlier quoted context omitted.

Open weight and open source are not the same

This is a pretty banal comment at this point. Open source is the term used in the LLM community. It's common and understood. Nobody is going to release petabytes of copyrighted training data, so the distinction between open source vs weights is a rather pointless one.

Tell this to the Allen project, Apertus Project, SmoLLM, etc, etc, etc

Re: DeepSeek v4

#713

Post by kittikitti »

This is a great model from DeepSeek and I look forward to seeing the developments from this. I am also very frustrated that American states, corporations, and organizations have banned DeepSeek models or made them illegal. It considerably restricts my AI operations and the ability to conduct research and development. As someone who hosts open-source models with compute resources available to serve DeepSeek V4, it brings considerable risk just because I am in America.

I hope that DeepSeek wins the AI race or at least gets ahead to the point where it becomes infeasible for bans and regulations against it. It's ridiculous that American legislators are advocating for less regulations for DeepSeek except for their own racist ideas about which AI should be approved or not.

Re: DeepSeek v4

#714

Post by wolttam »

I'm impressed! I've been giving the various open-weight models a particularly gnarly (for my brain, at least) refactoring/cleanup task in my DIY coding harness[0] - essentially, de-spaghettifi the main chat view's update logic, which had grown organically since early 2024.

Kimi 2.6 went hard and left me with a buggy mess. GLM 5.1 hedged and made a 25 line change (but it was an improvement). DS V4 went hard, fixed its issues along the way, and left me with a significantly nicer codebase! (...that I will now be spending some time testing before releasing to the project)

[0]: lmcli (simple, Go, nice UX, MIT licensed, works well with DS V4) https://codeberg.org/mlow/lmcli

Re: DeepSeek v4

#715

Post by SkyBelow »

gommm wrote: post #424
kitd wrote: post #386

Earlier quoted context omitted.

I can't find any info on what exactly is open sourced. And in any case what does open source actually mean for an llm? It's not like you can look inside it to see what it's doing.

For me open source means that the entire training data is open sourced as well as the code used for training it otherwise it's open weight. You can run it where you like but it's a black box. Nomic's models are good example of opensource.

Even with all training data provided, won't it still be a black box? Unless one trains it exactly the same, in the exact same order for each piece of data, potentially requiring the exact same hardware with specific optimizations disabled due to race conditions, etc., the final weights will be different, and so knowing if the original weights actually contain anything extra still leaves any released weights as a black box, no? There isn't an equivalent of reproducible builds for LLM weights, even if all of this was provided, right?

Re: DeepSeek v4

#716

Post by noah_buddy »

rfrey wrote: post #653
Ladioss wrote: post #376

Earlier quoted context omitted.

I always find it an illuminating experience about the power of mass propaganda every time I see an American believe they somewhat have the moral high ground over China, despite starting a new war somewhere around the globe either for petrol or on behalf of Israel every six months.

Many of us (worldwide, I'm not American) watched China massacre thousands of its own children at Tiananmen Square. The US is descending into totalitarianism, but it hasn't reached that level yet. And China may have changed in some ways but there have been no signals it would not repeat that event if it thought circumstances warranted.

Don’t you think that it’s a signal that the last major event you can point to is decades old?

Others may say “what about Uighurs?” or “what about Hong Kong?” but I think that the rest of the world is not doing all that much better on terms of civil repression.

In the UK, you can be arrested for voicing disagreement with the rationale for another person’s arrest (not generally, but on a specific hot button issue they’d rather not anyone talk about). French politicians are attempting to make illegal criticism of Israel, carte blanche. Don’t even get me started on Germany, which is so self-shamed from the last century they have overcorrected into legitimating an external state above all else. Across the pond, you hardly even have to convince anyone that it’s on the downtrend, unless they’re 30% of the population who believe the Don is christ alive (but don’t like if he says it).

The world is very unstable at this point and China is a country that strongly values and incentivizes stability, at the expense of individual rights. This is contra a lot of the west which is both unstable and actively undermining individual rights.

Re: DeepSeek v4

#717

Post by alecco »

patshead wrote: post #189
alecco wrote: post #117

Earlier quoted context omitted.

Please don't slander the most open AI company in the world. Even more open than some non-profit labs from universities. DeepSeek is famous for publishing everything . They might take a bit to publish source code but it's almost always there. And their papers are extremely pro-social to help the broader open AI community. This is why they struggle getting funded because investors hate openness. And in China they strug…

DeepSeek's models are indeed open weight. Why do you feel that pointing this out would be considered slander?

>> Truly open source coming from China.

> Open weight!

They clearly were implying it's not open source.

Re: DeepSeek v4

#718

Post by mchusma »

At first, I was more excited about the Flash model, but I'm now more excited about the Pro model in many ways. I feel like the Pro model with an Run through unsloth, and with some fine tuning, is gonna be enough for many vertical SaaS applications.

Where previously I was wary to under-provide the intelligence level, I'm now more excited about the idea of being able to give these pretty large intelligent models to my application. The idea that for basically sub-agents, we can fine-tune them, should reasonably expect to perform as well as Opus for a specific subtask of which my applications have many.

In other words, we can run a general-purpose intelligent model, Sonnet or Opus, orchestrating a fleet of, let's say, 30 to 50 of these sub-agents that have been fine-tuned. By doing that, I can get very low pricing versus something that would have occurred if I used Opus or Sonnet for everything.

Re: DeepSeek v4

#719

Post by solenoid0937 »

dzonga wrote: post #521
jari_mustonen wrote: post #239

Open Source as it gets in this space, top notch developer documentation, and prices insanely low, while delivering frontier model capabilities. So basically, this is from hackers to hackers. Loving it! Also, note that there's zero CUDA dependency. It runs entirely on Huawei chips. In other words, Chinese ecosystem has delivered a complete AI stack. Like it or not, that's a big news. But what's there not to like when…

Jensen Huang said this in his recent interview - that China has the best/most engineers, it has the chip making ability, it's a good thing they wanna build on a Nvidia stack - but if you push them they will build on an all Chinese stack - but the interviewer was being a numb head who kept parroting the propaganda of Western tech supremacy

> but if you push them they will build on an all Chinese stack

That's alright. It delays them at least.

Re: DeepSeek v4

#720

Post by glenstein »

Ladioss wrote: post #376
sigmoid10 wrote: post #332

Earlier quoted context omitted.

Still not sure how I feel about China of all places to control the only alternative AI stack, but I guess it's better than leaving everything to the US alone. If China ever feels emboldened enough to go for Taiwan and the US descends into complete chaos, the rest of the world running on AI will be at the mercy of authoritarian regimes. At the very least you can be sure noone is in this for the good of the people anym…

I always find it an illuminating experience about the power of mass propaganda every time I see an American believe they somewhat have the moral high ground over China, despite starting a new war somewhere around the globe either for petrol or on behalf of Israel every six months.

And by contrast what I find stunning is the inability to engage in meaningful comparative analysis of relative harms. There's a lot of spectacularly insightful attention to detail in so far as it mobilizes what aboutism arguments and then that attention mysteriously falls away when we ask questions like the extent to which these sides allow free press or democratic elections with multiple parties or permit fair trials. You used to not have to explain these things.
Post reply on HN