DeepSeek-V4-Flash Update • Hacker News Board

Re: DeepSeek-V4-Flash Update

#281

Post by 1234letshaveatw » Fri, Jul 31, 2026, 5:42 PM UTC

bwfan123 wrote: post #255

1234letshaveatw wrote: post #215

Earlier quoted context omitted.

That isn't the Chinese way. They are much more focused on undermine and extinguish. Just look at the European car industry- on its way to being non-existent after the market was flooded, bye bye manufacturing base. Undercut the US AI providers and wait them out, they will go private after they have a stranglehold

> Undercut the US AI providers and wait them out You could have said the same of linux - that it extinguished proprietary OSs at least for server usecases.

true, but the motivations of OSS contributors are benign (overall)

Re: DeepSeek-V4-Flash Update

#282

Post by dotancohen » Fri, Jul 31, 2026, 5:42 PM UTC

flexagoon wrote: post #85

markasoftware wrote: post #77

Earlier quoted context omitted.

Have you completed the identity verification? It's much more lenient once you have

Weird, I'm using 5.6 Sol through Chinese resellers and it reverse engineers stuff just fine

Do you have any suggestions for resellers? My Gmail username is the same as my HN username if you're not comfortable posting that here.

Thanks.

Re: DeepSeek-V4-Flash Update

#283

Post by gr_norm » Fri, Jul 31, 2026, 5:44 PM UTC

wkcheng wrote: post #26

If the benchmarks are real and reflect actual use, then this is an insane model. This 300B model outperforms the previous DS4 Pro preview model (1.8T params), and it looks like it outperforms GPT 5.6 Luna too. And it's still cheaper than Luna, even with the price decrease. Crazy.

OpenAI must've known this was coming, hence the Luna price drop. This competition is amazing!

Re: DeepSeek-V4-Flash Update

#284

Post by dotancohen » Fri, Jul 31, 2026, 5:44 PM UTC

realusername wrote: post #79

embedding-shape wrote: post #56

Earlier quoted context omitted.

> Also, it will never complain about security guards, I've been using it to reverse engineer binaries. Maybe I'm using too weak language in my prompts, but none of the OpenAI models I've used via codex has refused to reverse engineer binaries, is it supposed to? I'm sitting right now reverse-engineering a 3rd party firmware together with Codex and haven't hit a single guardrail. Meanwhile, I see people complaining ab…

I got an account warning on OpenAI (waved after I complained) just because I was asking it how to root some >10 years old Android device.

Sonnet recently even suggested I root a three year old device and provided instructions. I suppose that is depends on use case. My use case was exporting data from an abandoned application running on an S24 Ultra.

The idea was that I could continue to use the application in the future and export the new data, not that I would be able to recover the extent data already in there.

Re: DeepSeek-V4-Flash Update

#285

Post by OkGoDoIt » Fri, Jul 31, 2026, 5:46 PM UTC

flexagoon wrote: post #85

markasoftware wrote: post #77

Earlier quoted context omitted.

Have you completed the identity verification? It's much more lenient once you have

Weird, I'm using 5.6 Sol through Chinese resellers and it reverse engineers stuff just fine

I think I’m missing something, what do you mean Chinese resellers? As I understand it, it’s difficult to even access OpenAI in China, how could they be reselling it? Do you mean something like openrouter but a Chinese version or something?

Re: DeepSeek-V4-Flash Update

#286

Post by gr_norm » Fri, Jul 31, 2026, 5:48 PM UTC

fdsjgfklsfd wrote: post #256

siva7 wrote: post #232

Earlier quoted context omitted.

CCP will be happy! Go on and share all your data with them..

Yes, they trained on my open source code and Wikipedia edits and Stack Overflow answers that I shared freely, so it's only fair that they release their models as open source and share back to the community that created them. I only allow my training data go to open source models.

Yeah, at least when I share my training data with labs releasing their models openly (Chinese or American or otherwise) it's nominally so an even better open model will land in my hands in the future.

Re: DeepSeek-V4-Flash Update

#287

Post by namuol » Fri, Jul 31, 2026, 5:51 PM UTC

Can someone please explain how these models aren’t just fine tuned for benchmarks? I’m not plugged in to this space much but it seems like such an obvious problem…

Re: DeepSeek-V4-Flash Update

#288

Post by sroerick » Fri, Jul 31, 2026, 5:54 PM UTC

namuol wrote: post #287

Can someone please explain how these models aren’t just fine tuned for benchmarks? I’m not plugged in to this space much but it seems like such an obvious problem…

They definitely are - but also people are using them pretty extensively for work. So ultimately you can't really fake "is it good". But there's no real measurements of that when a model is released, so we are stuck with benchmarks.

Re: DeepSeek-V4-Flash Update

#289

Post by codemk8 » Fri, Jul 31, 2026, 5:58 PM UTC

applicative wrote: post #184

dannyw wrote: post #91

Deepseek and moonshot are the only two providers I consent to training for.

Xi going to shut down open-weighting of them in a matter of months. No one seriously doubts this. They will be too powerful and they will be gone.

Me.

Re: DeepSeek-V4-Flash Update

#290

Post by wg0 » Fri, Jul 31, 2026, 6:01 PM UTC

Don't know about the bench marks but I am getting Opus 4.7 level performance at fraction of cost with DeepSeek V4 Flash set to high. It is a reliable workhorse.