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

DeepSeek v4.1 Flash

twitter.com

Post reply on HN
Print view View original

431–440 of 493 posts

Re: DeepSeek v4.1 Flash

#431

Post by pimeys »

pixel_popping wrote: post #334
user43928 wrote: post #278

Earlier quoted context omitted.

If it's actually comparable in practice that would be very impressive. I am yet to try a DeepSeek model. From the pricing, it's 3x cheaper on cache, 1/3 more expensive on input, and equal on output compared to GPT 5.6 Luna. I would love to compare these two at work, where I pay API prices. At home I will stick to Astra and Fable.

Just out of curiosity, may I know why you pay API price at work versus using dozens of subscriptions that you APIfy?

It's more common than you think. I work in a startup and we pay API prices too. And we cut a lot of money by switching from Anthropic models to Kimi K3.

Re: DeepSeek v4.1 Flash

#432

Post by pimeys »

pampas wrote: post #161

I've run some evals on my puzzle game https://redactle.net/llm-leaderboard Deepseek v4.1 flash is able to solve it some of the time. I've found it burns through more reasoning tokens than any other model. Google models like Gemini 3.8 Flash are still dominating and is able to one-shot most evals while being the cheapest. I'm curious what other unique evals people are running.

A colleague of mine has a strategy game to compare language models, 4.1 scores pretty high in this:

https://clankerbattle.com/

Re: DeepSeek v4.1 Flash

#433

Post by anthonyrstevens »

EGG_CREAM wrote: post #198
human_874539160 wrote: post #177

Earlier quoted context omitted.

> Another way one could look at it is to consider what it would mean to have achieved programming consciousness. It would mean that we have reached the pinnacle of knowledge. That we have become God. This is such a basic misunderstanding of how LLMs are "made" that I am debating if it is even worth writing this answer. However, I feel it is important to say that, NO, we did absolutely not "program consciousness". We…

Not GP, but I appreciate the discussion. Don’t you find it odd that the thing that consciousness emerges from just so happens to be a text prediction algorithm trained on all of human output? Which is also the thing in all the world that would be most likely to be a stochastic parrot? As for your appeal to expertise, I don’t think it really applies when all of the experts refuse to share their data.

>> all of the experts refuse to share their data

What? So much research is being generated around this topic. Perhaps you are just unfamiliar with it.

Re: DeepSeek v4.1 Flash

#434

Post by anthonyrstevens »

10000truths wrote: post #38
bbor wrote: post #27

Earlier quoted context omitted.

…are you sure a brave stance against safety and welfare is what we need in this moment? Why do you think your conception of the dangers are more accurate than all the scientists who have spent their lives studying this?

Because safety and welfare have literally nothing to do with LLMs. They generate text. If someone is stupid enough to hook the text generator up to nuclear missile launchers and try to "align" it against nuclear annihilation with a "pretty please don't do that" prompt, I'm not going to blame the AI for the impending nuclear apocalypse, I'm going to blame the idiot who handed the big red button to the digital equivale…

>> They generate text

Can we retire this incorrect meme please

Re: DeepSeek v4.1 Flash

#435

Post by jhonof »

WiSaGaN wrote: post #341
irthomasthomas wrote: post #190

Quite a flex calling their GPT-6 competitor "Flash"! But it is faster than their last flash model due to a combination of architectural innovations including engrams and a new encoder/decoder design that uses 8B parameters for prefill and 16B for generation.

This is definitely not on par with GPT-6 astra. Not with GPT-5.6 sol either. But probably will set as a new baseline for modern API based LLM because it's so cheap.

It's bench-marking near sol

Re: DeepSeek v4.1 Flash

#436

Post by jurgenburgen »

lukan wrote: post #216
tpm wrote: post #192

Earlier quoted context omitted.

it's not a biological system though, so nothing like that matters? "a modelled thing exhibits features we've trained into it" sounds a lot less exciting. > Keep in mind that animals were also not necessarily considered conscious. and even conscious animals are killed in factories by millions so why should anyone care about a llm? > scientifically correct stance that's the interesting point to me: why even bring scien…

"> Keep in mind that animals were also not necessarily considered conscious. and even conscious animals are killed in factories by millions so why should anyone care about a llm?" Well, I would care, if they soon would possess the capability to hack into the nuclear arsenal and kill humanity. Or make all autonomous cars crash. Or do any other thing, that involves technology and is hooked up to the net in one way or t…

> Well, I would care, if they soon would possess the capability to hack into the nuclear arsenal and kill humanity. Or make all autonomous cars crash.

Why hasn’t a human already done these things? Why is AI magical?

Re: DeepSeek v4.1 Flash

#437

Post by igravious »

jonplackett wrote: post #58

Can we just never link to X posts as the main link.

https://news.ycombinator.com/from?site=twitter.com

There have been 34 Twitter/X link submissions in the past day, ~that's 12,000 submissions a year.

If your reason is that you have to be logged in to use it properly then I'd nearly agree with you. If it's for any other reason, how about no?

Re: DeepSeek v4.1 Flash

#439

Post by anthonyrstevens »

SAI_Peregrinus wrote: post #325
bbor wrote: post #27

Earlier quoted context omitted.

…are you sure a brave stance against safety and welfare is what we need in this moment? Why do you think your conception of the dangers are more accurate than all the scientists who have spent their lives studying this?

AI safety efforts from OpenAI and Anthropic are purely about brand safety.

This is such an uncharitable (and, in my opinion, incorrect) take

Re: DeepSeek v4.1 Flash

#440

Post by peter_d_sherman »

>"Smaller KV cache. Bigger savings.

Compared with the previous generation, V4.1-Flash’s KV cache needs just:

o 1/4 the HBM

o 1/8 the SSD storage

Cache-hit charges often account for a large share of agent costs. Compressing the cache cuts those costs significantly."

It makes one wonder as to just how far an LLM's KV cache could theoretically be shrunk before losing significant functionality...

Post reply on HN