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

481–490 of 1001 posts

Re: DeepSeek v4

#481

Post by yanhangyhy »

somehow i canot open the link. but in their chinese version's release article, in the end ,there is a quote from xunzi(https://en.wikipedia.org/wiki/Xunzi_(philosopher))

"Not seduced by praise, not terrified by slander; following the Way in one's conduct, and rectifying oneself with dignity." (不诱于誉，不恐于诽，率道而行，端然正己)

(It is mainly used to express the way a Confucian gentleman conducts himself in the world. It reminds me of an interview I once watched with an American politician, who said that, at its core, China is still governed through a Confucian meritocratic elite system. It seems some things have never really changed.

In some respects, Liang Wenfeng can be compared to Linux. The political parallel here is that the advantages of rational authoritarianism are often overlooked because of the constraints imposed by modern democratic systems. )

Re: DeepSeek v4

#482

Post by swiftcoder »

rohanm93 wrote: post #440
swiftcoder wrote: post #356

Earlier quoted context omitted.

> For context, for an agent we're working on, we're using 5-mini, which is $2/1m tokens. This is $0.30/1m tokens. And it's Opus 4.6 level - this can't be real. It's doesn't seem all that out there compared to the other Chinese model price/performance? Kimi2.6 is cheaper even than this, and is pretty close in performance

Kimi is indeed somewhat cheap for frontier-level intelligence, but still is $4-5 per mm tokens. Deep Seek is at least an order of magnitude cheaper.

Oh, right you are. I misread where the decimal place was in the Deepseek pricing. That is incredibly cheap

Re: DeepSeek v4

#484

Post by Grp1 »

DeepSeek’s docs say V4 has a 1M context length. Is that actually usable in practice, or just the model/API limit?

Codex shows ~258k for me and Claude Code often shows ~200k, so I’m curious how DeepSeek is exposing such a large window.

Re: DeepSeek v4

#485

Post by lucrbvi »

Grp1 wrote: post #484

DeepSeek’s docs say V4 has a 1M context length. Is that actually usable in practice, or just the model/API limit? Codex shows ~258k for me and Claude Code often shows ~200k, so I’m curious how DeepSeek is exposing such a large window.

They have added a lot of optimization focussing on the KV-cache, so they can have a much larger window without eating all the VRAM.

The 1M window might be usable, but it will probably underperform against a smaller window of course.

Re: DeepSeek v4

#486

Post by adammarples »

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.

Yes the weights are basically compiled code, compiled from the source data and the training code.

Re: DeepSeek v4

#487

Post by samrus »

teiferer wrote: post #373
spiderfarmer wrote: post #290

Earlier quoted context omitted.

Just ask it for a summary of the USA’s role in Iran, Gaza, Lebanon and its recent threats against Panama, Cuba and Greenland! It might be able to keep track.

Does all this insane behavior from the US justify the Chinese censorship?

Of course not. But its disengenuous to only mention one like the US is clearly th lesser of 2 evils

Re: DeepSeek v4

#488

Post by gunalx »

avereveard wrote: post #193
0xbadcafebee wrote: post #70

Earlier quoted context omitted.

I don't think we need to compare models to Opus anymore. Opus users don't care about other models, as they're convinced Opus will be better forever. And non-Opus users don't want the expense, lock-in or limits. As a non-Opus user, I'll continue to use the cheapest fastest models that get my job done, which (for me anyway) is still MiniMax M2.5. I occasionally try a newer, more expensive model, and I get the same resu…

eh idk. until yesterday opus was the one that got spatial reasoning right (had to do some head pose stuff, neither glm 5.1 nor codex 5.3 could "get" it) and codex 5.3 was my champion at making UX work. So while I agree mixed model is the way to go, opus is still my workhorse.

I find gemini pretty good ob spatial reasoning.

Re: DeepSeek v4

#489

Post by kiproping »

thefounder wrote: post #480

They still don’t support json schema or batch api. It’s like deepseek does not want to make money

What do you currently use for json and batch, I was doing some analysis and my results show that gpt-oss-120b (non batch via openrotuer) is the best for now for my use case, better than gemini-flash models (batch on google). How is your experience?

Re: DeepSeek v4

#490

Post by d3Xt3r »

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…

> Also, note that there's zero CUDA dependency.

So does this mean I can run this on AMD? And on a consumer 9000 series card?

Post reply on HN