Top
Best
New

Posted by tosh 14 hours ago

DeepSeek V4 Flash 0731(arcprize.org)
578 points | 348 commentspage 6
dcchambers 14 hours ago||
This latest DeepSeek is almost at the "too cheap to meter" level. That's going to be a larger unlock than models like Fable/Mythos that are way too expensive to justify, IMO.

What secret sauce do they have?

est 7 hours ago||
> What secret sauce do they have?

Quant company usually squeezing every penny.

pama 12 hours ago|||
No secrets—all published. Very efficient attention. Excellent kernels. Great caching subsystem. Small and well trained model.
throwaway_95283 13 hours ago||
limited resources, no modern GPUs, no $10 billion dev budgets.

pair it with codewhale, 50 agents, 200 MB of ram.

m3kw9 8 hours ago||
The token price seem to be jigged, how do you know if it's subsidized or temporary. Anyone can just lower the token price to get to the left.
jacquesm 5 hours ago|
Almost every service provider in the AI field is subsidizing their token cost to some degree, they're all shooting for marketshare and lock-in (and they're not really achieving the latter).
iagooar 13 hours ago||
I love DeepSeek V4 Flash since the pre-0731, now even more. It is the first model that is truly too cheap to meter.

But I find it having a pretty significant problem with tool calling - no idea why, but tool calling with it is SLOW. As long as the model is reasoning, all good. But give it a bunch of tools and it becomes extremely slow.

Am I the only one experiencing this?

system2 9 hours ago||
How can OpenAI or Anthropic fight against these prices?! $0.14 input, $0.28 output. For 1M tokens...
gxs 9 hours ago||
One thing that popped into my head is that this shows how committed they are to building something that scales across the world

China has zero energy concerns in terms of energy production - not literally zero, but they’d be able to prioritize other dimensions and not necessarily worry about efficiency

Here they are though releasing models that sip resources

casey2 11 hours ago||
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.
esafak 14 hours ago||
It's serviceable but, like many Chinese models, it uses a lot of tokens to get work done.
gruez 13 hours ago||
>it uses a lot of tokens to get work done.

That's irrelevant when you use $/task as the metric, which the OP does use.

esafak 13 hours ago||
It also affects the time.
aitchnyu 12 hours ago|||
It felt like a rocket compared to GLM 5.2 though. Are Chinese models generally token-heavy?
esafak 12 hours ago||
https://artificialanalysis.ai/?cost=intelligence-vs-cost-per...
cyanydeez 12 hours ago||
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

nomel 10 hours ago||
> 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.

cyanydeez 8 hours ago||
It's specific to the harness. Using dynamic context pruning, the budget cuts it off after a select amount of tokens and the budget message tells the model to use subgents to finish whatever it's thinking about
nomel 8 hours ago||
Nice. Does it use a summarization, or a hard cutoff?
ttkciar 7 hours ago||
llama.cpp uses a hard cutoff. The agent then does "something" that is specific to the agent's implementation and configuration. It might summarize and then "finish the thought" with a different model, and then resubmit the prompt to the llama.cpp API endpoint with <think>..</think> prefilled. The primary model then infers the remainder of the reply.
leizhou 13 hours ago||
so cool. does it mean it can understand the verificated code
WhitneyLand 13 hours ago|
The DeepSeek team is so strong, very impressive.

Imagine if they had GPU resources of western labs.

mosura 13 hours ago|
Necessity is the mother of invention.

SV companies get way too comfortable when they have enough in the bank to stay running more than three months.

More comments...