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

DeepSeek V4 – almost on the frontier

simonwillison.net

Post reply on HN
Print view View original

31–40 of 420 posts

Re: DeepSeek V4 – almost on the frontier

#31

Post by baldai »

wg0 wrote: post #9

Deepseek v4 Pro feels like Claude Opus 4.6 in it's personality but here's what I did find out about costs: I did cut loose Deepseek v4 on a decent sized Typescript codebase and asked it to only focus on a single endpoint and go in depth on it layer by layer (API, DTOs, service, database models) and form a complete picture of types involved and introduced and ensure no adhoc types are being introduced. It developed a…

Only similarity it has to Opus 4.6 is the 4 in the name. I do not understand these dishonest comparisons. OOS models are vool, cheap and promising for a future -- but why are we pretending they are better than they are?

Re: DeepSeek V4 – almost on the frontier

#32

Post by notrealyme123 »

deaux wrote: post #27
gmerc wrote: post #25

Earlier quoted context omitted.

Because they give it away for free and offer APIs at very acceptable rates. Not that hard to figure out, Robin Hood stealing our data tax back comes to mind.

GitHub is free.

User publishes to github => Copilot trains with GitHub data => MS Sells copilot => User workes for Microsoft (in the sense of giving it's labour for MS to make money)

User publishes to github => Deepseek trains with GitHub data => Deepseek gives model away for free => User did not work for Deepseek (in the sense of giving it's labour for Deepseek to make money)

Re: DeepSeek V4 – almost on the frontier

#33

Post by raincole »

deaux wrote: post #14

I'm surprised that people here don't care at all about these models openly training on your data, especially if you use them straight from the model developer. Whereas things like "GitHub now automatically opts everyone into using their code for model training" get hundreds of justifiably angry comments, I never see this brought up anymore on posts like these talking about using Chinese models through OpenRouter. Thi…

Two factors. First is anti-americanism (or at least anti-american-capitalism).

But the more important one is the social contract. Github came far before LLM era. The branding around it is being the storage of open source projects and many users want to it stay away from AI hype. You won't expect LLM providers to stay away from AI hype (duh) so it's less an issue for them.

Re: DeepSeek V4 – almost on the frontier

#34

Post by notrealyme123 »

prism56 wrote: post #18
deaux wrote: post #14

I'm surprised that people here don't care at all about these models openly training on your data, especially if you use them straight from the model developer. Whereas things like "GitHub now automatically opts everyone into using their code for model training" get hundreds of justifiably angry comments, I never see this brought up anymore on posts like these talking about using Chinese models through OpenRouter. Thi…

If the data is opensource on github, then in my opinion it should be fair game.

Things being public should not be enough. just because someone leaked your medical information to the public via a data breach should not make it fair game. There should be some rules.

Re: DeepSeek V4 – almost on the frontier

#36

Post by amelius »

ithkuil wrote: post #23
wg0 wrote: post #9

Deepseek v4 Pro feels like Claude Opus 4.6 in it's personality but here's what I did find out about costs: I did cut loose Deepseek v4 on a decent sized Typescript codebase and asked it to only focus on a single endpoint and go in depth on it layer by layer (API, DTOs, service, database models) and form a complete picture of types involved and introduced and ensure no adhoc types are being introduced. It developed a…

Even taking into account the fact that they are billing at 75% discount it's still quite cheaper

Aren't they all billing at discount?

Re: DeepSeek V4 – almost on the frontier

#38

Post by ForHackernews »

ozgrakkurt wrote: post #24
prism56 wrote: post #18

Earlier quoted context omitted.

If the data is opensource on github, then in my opinion it should be fair game.

IMO this is unfair for GPL or similarly licensed code. Seems ok for MIT like licensed code though

It's totally fair to use GPL code, it just means all the models built by Anthropic, OpenAI, etc. using GPL-licensed source are themselves bound by the GPL. Plus, any works created downstream using those AI tools.

We're on the verge of a golden age of software as soon as someone finds a court with courage.

Re: DeepSeek V4 – almost on the frontier

#39

Post by zozbot234 »

raincole wrote: post #35

The V3/R1 time and now are in such contrast. V3/R1 were hyped hard and barely usable for coding. V4 is much less hyped but (anecdotally) it has completely demolished all the Flash/Lite/Spark models.

Huh? R1 was one of the earliest openly available MoE and reasoning models, that's definitely not "hype". People tried to do reasoning before by asking the model to "think it through step by step" but that was a hack. The later V3.1 and V3.2 releases AIUI unified reasoning/non-reasoning use under a single model.

Re: DeepSeek V4 – almost on the frontier

#40

Post by prism56 »

notrealyme123 wrote: post #34
prism56 wrote: post #18

Earlier quoted context omitted.

If the data is opensource on github, then in my opinion it should be fair game.

Things being public should not be enough. just because someone leaked your medical information to the public via a data breach should not make it fair game. There should be some rules.

I feel that's a flase dichotomy. The code visible on github is freely available for anyone to read and learn from.
Post reply on HN