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

DeepSeek V4 Flash 0731

arcprize.org

Post reply on HN
Print view View original

191–200 of 477 posts

Re: DeepSeek V4 Flash 0731

#191

Post by the_duke »

nylonstrung wrote: post #173

Compared to the last Deepseek V4 Flash version I've had tons of issues with it getting in infinite loops and talking to itself without executing tool calls, wasting tons of tokens This is on Pi agent, nothing fancy at all about my prompts or use case. Anyone else experiencing this? I've also had it randomly go from talking about Rust to talking about the electric chair, controversies about D&D rules (both irrelevant…

Yeah, I saw the same thing - quite annoying. It can be mitigated through the prompt.

Re: DeepSeek V4 Flash 0731

#192

Post by albedoa »

paxys wrote: post #118
LaurensBER wrote: post #9

I've been using it extensively since the release and the best summary I can give is that it's good enough to use it for (almost) everything and cheap enough that the cost are irrelevant. I'm running it in Oh My Pi with a second instance running as "advisor" and even with 5-6 active sessions (effectively 12 streams) I'm struggling to spend more than 5 bucks per day. OpenCode Go even has double limits temporarily so fo…

How is $5/day irrelevant? In the $150/mo range you can get effectively unlimited usage of GPT 5.6 Sol (Pro plan). Why use a much weaker model for the same price?

Ignoring the other side of the equation is a pretty wild thing for you to do here:

> I'm running it in Oh My Pi with a second instance running as "advisor" and even with 5-6 active sessions (effectively 12 streams)

Re: DeepSeek V4 Flash 0731

#193

Post by casey2 »

Finally something that is breaking away from the pack. Interesting that max costs less than high. I still think, currently, TPS is more important than near frontier intelligence. Likely for reasons that LeCun outlined, maybe out of a billion prompts you will get value from that intelligence. When we have very fast models abstraction will work as that filter.

Re: DeepSeek V4 Flash 0731

#194

Post by FernandoTN »

abixb wrote: post #162
LaurensBER wrote: post #9

I've been using it extensively since the release and the best summary I can give is that it's good enough to use it for (almost) everything and cheap enough that the cost are irrelevant. I'm running it in Oh My Pi with a second instance running as "advisor" and even with 5-6 active sessions (effectively 12 streams) I'm struggling to spend more than 5 bucks per day. OpenCode Go even has double limits temporarily so fo…

If what you're saying is true and accurate, then US-based AI labs are in big trouble. The only saving grace might be some sort of a 'national security' proclamation banning the use of state-of-the-art Chinese (and non-US) models across US federal and state governments and large enterprises (especially ones with federal government contracts), but even still, US AI labs will probably lose out massively on international…

I think you're overlooking the fact that for long-horizon tasks, even small errors compound over time and can lead to catastrophic outcomes.

For simple queries, we have reached the threshold since the beginning of the year, and models are good enough from every provider to make a meaningful difference between one another. (ChatGPT, Claude, Gemini, Grok, MuseSpark, Kimi, DeepSeek, GLM...)

The real unlock will be, and you can already see it with GPT-5.6 and Fable-5, to delegate complex enough tasks that will take more than 24 hours to get done and they will not lose track. I'm not talking about a loop, but the actual intelligence to recover from these compounding errors that accumulate in dumber models.

We're still a long way from the intelligence needed to let one of these agents go ahead and supervise multiple layers of sub-agents underneath to do complex orchestration. The future looks very promising and exciting. Imagine having the possibility of a Frontier model orchestrating as many sub-agents as needed that are running on cheaper models like DeepSeek.

Re: DeepSeek V4 Flash 0731

#195

Post by usef- »

_aavaa_ wrote: post #134
paxys wrote: post #118

Earlier quoted context omitted.

How is $5/day irrelevant? In the $150/mo range you can get effectively unlimited usage of GPT 5.6 Sol (Pro plan). Why use a much weaker model for the same price?

I really doubt that the either of the pro plans are subsidized heavily enough to support you swapping v4Flash for Terra, much less Sol. $5/days is ~330 Mtok/day, that’s a nontrivial amount of work, and none of the gpts are more efficient than deepseek at $/task if deepseek meets your quality bar.

They do give a significant amount more than you would expect from the API pricing. The US provider API pricing has heavy margins by all accounts (and most are short enough of GPUs that there's little incentive to drop).

Re: DeepSeek V4 Flash 0731

#196

Post by embedding-shape »

metadat wrote: post #19
Aeolun wrote: post #15

Earlier quoted context omitted.

But DeepSeek now has a warning they’re going to sharply increase their API pricing sometime in the future.

Source?

I made a HN submission about it yesterday with a copy of the email: https://news.ycombinator.com/item?id=49197176

Re: DeepSeek V4 Flash 0731

#197

Post by minraws »

minraws wrote: post #45
LaurensBER wrote: post #21

Earlier quoted context omitted.

Dax (from Opencode) has tweeted that they can replicate or beat the price with rented GPUs. Deepseeks secret sauce is the incredibly cheap caching (magnitude cheaper than other providers). vLLM has recently released a similar approach. It's not as effective as what DeepSeek does but still an interesting development. I have no doubt that in due time other providers will match or perhaps even beat the current DeepSeek…

As someone who recently tried it on some blackwell cards, it's possible to match the prices especially the input can be even cheaper and output can match the costs so you can easily build a net 20-30% margin business even at current GPU prices. The entire issue is caching, I tried to write some custom to dump to disk kv-caching using some ideas from their papers and my experience with snapshots and vm checkpoint syst…

Mild info dump, since this has a few too many upvotes and some folks might be misunderstanding, 20-30% is assuming a typical agentic workload where input tokens dominate by over 20:1 or at least 10:1, if you are output token heavy then this is going to be a different ball game.

And there is no way in hell anyone can afford caching prices same as what DeepSeek is offering, and DeepSeek keeps the cache available for an insane amount of time most providers will flush it in 5-mins like Claude/Anthropic (some offer customizing it but I am not sure of the pricing, it's load based on some like Fireworks, which means assume a couple minutes at most, they say several minutes god knows what that really means).

There is no way to match DeepSeek's current prices, "profitably" if you are renting a GPU and reselling tokens, unless you have some really amazing caching infra or something.

Deepseek's prices are just insanely cheap, I am not saying it's impossible to get there the overall performance suggests it should be feasible, but I will be damned if any provider could match their tps and caching any time soon at those same prices profitably.

I believe even if Deepseek 2-3x their prices across the board even then they would be cheaper for most long running tasks, that's just how good their caching is.

For one I have managed to hit the cache after over 24 hours on their system it's insane, I honestly didn't care because it was so cheap but it truly made me incredibly happy to think about the engineering that must have taken. TTFT is slightly worse, but it's good enough, for those cache prices I can take a few seconds worth of hit on TTFT.

Re: DeepSeek V4 Flash 0731

#198

Post by nomel »

cyanydeez wrote: post #128
esafak wrote: post #17

It's serviceable but, like many Chinese models, it uses a lot of tokens to get work done.

If I had the GPU size, hook it up to llama.cpp and setup the --reasoning-budget and reasoning-message; Most of that additional reasoning is a lot of garbage and you can redirect it to useful output. That's how I handle the Qwen27B and 35B

> Most of that additional reasoning is a lot of garbage and you can redirect it to useful output.

What do you mean by "redirect it to useful output"? Could you give an example? This sounds interesting.

Re: DeepSeek V4 Flash 0731

#199

Post by podnami »

rpdillon wrote: post #113
amelius wrote: post #88

Earlier quoted context omitted.

> it's good enough to use it for (almost) everything which in your case is?

I've posted a few times about my project that's a collection of 30k-250k webapps that are served from a WebDAV server. The apps know how to write updated copies of themselves back to the server. My family uses it. I have gallery apps (yearbooks for each year are a lot of fun!) of us on trips and just living, an outlining app that's a mesh of Workflowy and Org Mode (it's called Fluxtral), a markdown-backed app (it use…

A collection of 30k-250k apps? Like individual unique apps?
Post reply on HN