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

DeepSeek V4 Flash 0731 Intelligence, Performance and Price Analysis

artificialanalysis.ai

Post reply on HN
Print view View original

171–180 of 342 posts

Re: DeepSeek V4 Flash 0731 Intelligence, Performance and Price Analysis

#171

Post by rapatel0 »

Der_Einzige wrote: post #53
freakynit wrote: post #27

I will get downvoted, but fck it. The ban on these open models is coming within weeks, if not days. As usual, the excuse will be "national security".

Hell, the US doesn't even need to act. I claim the CCP will wise up within 2 years, possibly much much sooner, and ban their own companies from open sourcing to prevent the Americans from acquiring the capabilities. Despite all the nonsense claims of China distilling US models, the reality is that the Americans absolutely do distill these free Chinese models, and distillation when full logprobs are available (i.e. yo…

You’re describing the dump and pump strategy that china has historically used across a number of industries. Stands to reason that this is what is going on.

Re: DeepSeek V4 Flash 0731 Intelligence, Performance and Price Analysis

#173

Post by miyuru »

lima wrote: post #123
0cf8612b2e1e wrote: post #106

Somewhat relatedly, how do the economics for Huggingface work? They must be hosting petabytes of models and datasets by now. I have downloaded quite a few “just in case”, only to replace them with the later iteration months later. Does the file hosting actually cost peanuts when you do it yourself and the cloud has shattered my understanding of what it actually costs to deliver so much data?

Bandwidth is really cheap when you run your own infra

when trying to download models using browser, the DL domain for me us.aws.cdn.hf.co which points to bunch of singapore AWS IPv4 for me, despite it saying US.

Re: DeepSeek V4 Flash 0731 Intelligence, Performance and Price Analysis

#174

Post by donquichotte »

Saline9515 wrote: post #137
dpacmittal wrote: post #91

[flagged]

Good luck when the last non-Chinese frontier labs will have closed and the CCP will ask to stop sharing models open source.

My conspiracy theory is that this is the new space race, and the CCP encourages this to show the world what Chinese engineers are capable of, and tank the Anthropic/OpenAI valuation bubble as a desirable side effect.

Re: DeepSeek V4 Flash 0731 Intelligence, Performance and Price Analysis

#177

Post by wongarsu »

0cf8612b2e1e wrote: post #106

Somewhat relatedly, how do the economics for Huggingface work? They must be hosting petabytes of models and datasets by now. I have downloaded quite a few “just in case”, only to replace them with the later iteration months later. Does the file hosting actually cost peanuts when you do it yourself and the cloud has shattered my understanding of what it actually costs to deliver so much data?

File hosting is pretty cheap. Egress traffic is about $90/TB in the cloud, but around $1/TB in the real world. Storage is in the realm of $5/TB/month after adding redundancy

At the scale of Huggingface, that still amounts to a lot load of money. Significantly less than if you did the same in AWS, but still a lot

That said, they do have a deal with AWS to make the data available in AWS ip space. Maybe they got some cheap hosting out of that too

Re: DeepSeek V4 Flash 0731 Intelligence, Performance and Price Analysis

#178

Post by segmondy »

bwfan123 wrote: post #165

> For the Code Agent tasks among the public benchmarks above, DeepSeek-V4-Flash-0731 is evaluated with the minimal mode of DeepSeek Harness (to be released) as the agent framework So, are they planning to announce an optimized coding agent harness as well ? DSv4 flash is a fantastic model, and my daily driver. With reasonix or pi, I can code all day long and pay a few pennies for it. No token anxiety. Whereas the sam…

They announced it already, read the tech report.

"For the Code Agent tasks among the public benchmarks above, DeepSeek-V4-Flash-0731 is evaluated with the minimal mode of DeepSeek Harness (to be released) as the agent framework, using the max reasoning effort level with temperature = 1.0, top_p = 0.95."

Re: DeepSeek V4 Flash 0731 Intelligence, Performance and Price Analysis

#179

Post by smrtinsert »

I have been loving deepseek flash. I can code in my hitl preference doing complex but moderate amount of coding and not even hit 1 dollar in a session. It gets harder to justify paying 20/mo to Anthropic for unused compute when I have it on demand and at a much cheaper rate with deepseek.
Post reply on HN