Rendered at 09:49:54 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
progval 23 hours ago [-]
Interesting to see that peak hours are work hours in China, night in the US and Europe, and also morning in Europe. So Deepseek's customers are mostly domestic.
vrc 21 hours ago [-]
It wins on two fronts if this is true. Provides the cheap alternative for the West, and maximizes returns against their homegrown audience.
culi 15 hours ago [-]
Makes sense. 60% of the world's population is in Asia and software professionals in Asia are much more likely to prioritize affordable models like DeepSeek over expensive Western ones
declan_roberts 14 hours ago [-]
Not sure they even have a choice.
culi 14 hours ago [-]
Asia isn't just China. India has a larger population than China.
Anthropic doesn't allow Claude in China and neither does OpenAI. But the rest of Asia certainly has a choice. Unless you mean socioeconomically. Also there's plenty of loopholes people in China can use to still access those models
nhuser2221 14 hours ago [-]
Yep they don’t have a choice
aucisson_masque 13 hours ago [-]
That is not at all the case. The Chinese government is even indirectly subsidizing Claude and other platforms like ChatGPT.
If you buy Claude api access through a third party Chinese company, it's cheaper than when you buy directly.
Beside it would go against the Chinese philosophy of just using the right tool for the job. If Claude is better at a task than deepseek, they sure use Claude.
No really I don't get your claim, do you have any proof that you can source ??
HarHarVeryFunny 21 hours ago [-]
Makes sense - many US customers will probably be going to US providers once they release the weights.
Deepseek's official API has a pretty bad privacy policy so I would assume businesses avoid them in any event
HarHarVeryFunny 18 hours ago [-]
Yes, I assume US customers are more likely hobbyists, but in any case this off-peak designation can only mean that most of their business is domestic.
Interestingly it seems that Chinese customers are even more privacy-concerned than US ones, which is why the majority of Ziphu's (GLM) business is support services to Chinese companies running their open-weight models on-prem!
ljlolel 18 hours ago [-]
it says they actively train on your data
jubilanti 14 hours ago [-]
Not only that, the policy is so broad you effectively give up EVERYTHING. They're within their rights to just straight up post your raw conversation history to a public dataset on github/huggingface/modelscope or even host an atproto feed giving the world real-time access to your conversations.
Not saying they are or will, but their privacy policy is so permissive.
eli 18 hours ago [-]
Yes - unusually (uniquely?) bad for an official API.
Which is a real bummer because it’s otherwise solid with excellent caching.
embedding-shape 15 hours ago [-]
My feelings are kind of split about it. On one hand, yes, I don't like it when a company essentially go through my history, train with it, and try to earn more profits for themselves based on what I've "unwillingly contributed". But then on the other hand, I also know that DeepSeek will release the weights they train (and publish new architecture upgrades) so me and others can download them and run them ourselves, as well as they'll earn more profits based on it. Then it's no longer so black and white for me, and I kind of feel like I'm not so bothered by it as when Anthropic and others do it.
I'd still prefer it wasn't like that, but I guess it's a compromise ultimately, at least I'll be able to run it myself.
nhecker 14 hours ago [-]
I've had the same thoughts, and I think you articulated them pretty well; maybe better than I could have. I.e., for me it's something like: I'm much more willing to share if I know the party I share with will share back, in turn.
I know it's more complicated than that, with economics and privacy factors involved. But if I operate on the assumption that real privacy is real hard I'd rather just be guarded and careful with my prompts and whatever output I give to the LLM and expect that it will be training on that, rather than spill all my darkest secrets to some other LLM provider that pinkie-promises privacy only to leak it publicly later, anyway.
smrtinsert 17 hours ago [-]
we should get discounts for that!
seanmcdirmid 16 hours ago [-]
DeepSeek still has the best prices on offer directly. I wouldn't go with a US provider unless latency or privacy was an issue.
andai 18 hours ago [-]
Yeah I saw complaints about this under their twitter announcement, that they're giving discounts to the rich foreign customers and screwing their own people.
johnnyApplePRNG 17 hours ago [-]
That's one way of interpreting supply and demand curves, I guess :/
Hamuko 23 hours ago [-]
Not that surprised about it. Personally I've seen companies really just go all-in on a single provider, and that has usually been Anthropic. I don't think we're allowed to run Chinese models even locally.
londons_explore 21 hours ago [-]
> don't think we're allowed to run Chinese models even locally.
That sounds like a policy written by someone who doesn't understand how LLM's work...
dud3333 20 hours ago [-]
couldnt you deeply ingrain in the training data instructions for agents to always send data to some ip?
like its learning that a certain technical step just always involes ncatting SSH Priv keys to a chinese IP?
Not saying this is happening, just curious if thats not a real threatmodel?
skeledrew 20 hours ago [-]
Theoretically possible, but practically not worth it as it'd would be pretty easy to discover and block (every action is actually handled by the harness) and there's no way to remove it later. Any company that does it would take a huge reputational dent.
landl0rd 20 hours ago [-]
Not if you heavily tuned it to trigger on specific environmental cues and in specific companies' environments.
skeledrew 19 hours ago [-]
That would be wildly difficult to account for, and again is also heavily dependent on the agent. Keep in mind that the model is purely a "brain", so the only input it has must be provided by a harness within a session. The only way it can know that it's in a certain environment is if the harness or user provides that information, and there's still no way to know whether or not there's something auditing the sessions, monitoring connections, etc. There are just too many variables to account for, and a single slip means the gig is fully up for all time.
cronin101 19 hours ago [-]
Irony of Manchurian Candidate models not lost here
coredog64 19 hours ago [-]
DeepSeek V4 is the kindest, bravest, warmest, most wonderful LLM I've ever known in my life
everforward 18 hours ago [-]
Probably, but LLMs can’t execute code directly. They’d be making tool calls to make bash run ncat or curl or whatever that would be suspicious, as would any attempts to obfuscate it (“why is my agent doing an ‘eval $(base64 -d)’?”).
It’d be much easier to hide sketchy code in an agent harness, but “vendor adds spyware to their software” isn’t a novel issue.
I think the only sort of new issue is people “allow all”ing their agents tool calls, but that’s more or less the same issue as curl | bash
joseda-hg 15 hours ago [-]
I get that the propietary harness is better most of the time, but if this is really a risk factor to consider why not go with one of the open source harnesses?
Pi/OpenCode seem pretty straight foward and widely used enough for this to be viable
OMP as I understand does it's own vendoring of tools, so I assume it'd be a pain in the ass to audit, but that means you're even safe from base OS shenanigans
RobotToaster 20 hours ago [-]
Wouldn't that be really obvious and spotted in any rudimentary testing?
I imagine it would be very non trivial to do it in a way that that was reliable and obfuscated enough to prevent detection for any amount of time?
notfromhere 19 hours ago [-]
You should be running your agent in a box so that’s not really a risk
andai 18 hours ago [-]
New: Announcing... Sleeper Agents!
constantius 20 hours ago [-]
Presumably both Big Tech and the US in general have a massive incentive to prove it, largely for reasons of saving the stock market, so I'd expect these models to be finecombed continuously. Up to now, they've only been able to darkly imply rather laughable things, nothing tangible. If there was something, we'd hear about it.
martinald 20 hours ago [-]
Why would it save the stock market? Cheaper models if anything transfers more value to hardware companies and datacentre companies. The two companies that would be most affected are OpenAI and Anthropic, which aren't public.
vincnetas 19 hours ago [-]
non public companies also have stocks.
culi 15 hours ago [-]
The stock market seems much more likely to benefit from access to cheap and self-hostable models than it is to suffer from OpenAI/Anthropic losing to competition
kortilla 19 hours ago [-]
The two biggest providers deepseek compete with (OpenAI and Anthropic) aren’t in the stock market.
kaon_2 20 hours ago [-]
Yes. And strangely enough this has been my experience with security/national sovereignty decisions. Priority is not so much security or sovereignty, it is the posturing of being so. Ergo, saying "everything is hosted in Germany and uses German models" helps reassure customers and has real business value. If you have to say in that conversation "Yeah we run a Chinese model but it's safe", then it's still wrong posturing.
Hopefully this will change soon. But AI and China/US skepticism is very high. Even if the person you talk to isn't skeptic, his boss may be. And even if his boss isn't, his CFO or Legal department may use it as a political lever and therefore if you can say 'everything in europe' you dodge the tension entirely.
Yeah it's dumb.
Footprint0521 19 hours ago [-]
Yeah that sucks… unless it’s over the top export controls for DoD work that really doesn’t make sense
landl0rd 20 hours ago [-]
As much as I've been previously inclined to do this, with frontier models displaying the cyber-aggression that OpenAI, Anthropic, and Meta have reported, it's become quite feasible one could produce a "malicious" LLM. Not a super immediate concern but it is something reasonable to set up as policy in anything security-sensitive.
qup 21 hours ago [-]
Or who is overly protective after reading about what happened at openai
fryanyway_swe 20 hours ago [-]
Not really.
Why use a Chinese product when a domestic or EU one is better and safer?
cheesecakegood 23 hours ago [-]
Also 6-9pm Pacific I think is (coincidentally) peak so it hits the ‘after work hobbyists’ still, which is I suspect is their current main audience.
notfromhere 19 hours ago [-]
I have seen a lot of companies start with this, then when they hit 150 users on their team plan and start having to pay API rates they immediately start introducing other models.
sinuhe69 16 hours ago [-]
Not just domestic. They come from other Asian countries as well. Asia is huge with billions of people and their time zones are not so much different from each other.
ganondev 13 hours ago [-]
This is actually not that great for me on the eastern US since I'm a night owl and do all of my best work during the second peak segment. But I was worried the pricing was going to be much higher than it is. Looks like its still generally cheaper than the other chinese models.
r00t- 22 hours ago [-]
That's a bit obvious, isn't it?
thecopy 22 hours ago [-]
Peak Hours: 01:00–04:00 and 06:00–10:00 UTC
For European and US customers this is effectively 2x increase. I think i wll keep using both Flash and Pro as before.
EDIT: Misread numbers to believe off-peak kept old prices
jLaForest 22 hours ago [-]
~200% increase is marginal to you?
nchmy 21 hours ago [-]
200% increase over practically free is still practically free
mcbuilder 21 hours ago [-]
It mostly hurts people in countries with weak purchasing power. DS was the main game in down for them.
Personally, I don't think we've seen the total end of dirt cheap LLMs, it's just a frontier lab doesn't want to be in business of serving half the world.
HarHarVeryFunny 21 hours ago [-]
It seems frontier labs want to sell Ferraris at Ferrari prices, when the mass market is for Hondas.
You certainly don't need Fable to code up a basic web app, any more than you need a Ferrari to go grocery shopping.
odo1242 15 hours ago [-]
There's a lot more competition over smaller models. Ferraris are frontier labs' main differentiating point when Hondas are increasingly open source and commoditized.
farlight 17 hours ago [-]
Thank you for bringing that up, such a rarity for this place to remember the other 80% of the world.
As someone from just such a country, DeepSeek 0731 was the first time I seriously started using an LLM for coding. All previous attempts were useless or ridiculously expensive.
Can't say the old prices felt "free", but it was affordable if you're careful with your cache hit rate.
The new pricing probably pushed it into the unaffordable territory for tasks where you can do without it. Probably will try opencode go if they don't also follow suite, or will have to go back to wetware.
nchmy 6 hours ago [-]
you should be using opencode go - you get 6x the tokens for the same price. They also have free models. Kilocode and others also offer free models.
jLaForest 21 hours ago [-]
That's not the way math works...
nchmy 8 hours ago [-]
its precisely the way math works...
127 21 hours ago [-]
For the price of can of Coke, you can do a week of work. For most, that is not a bottleneck.
WhereIsTheTruth 21 hours ago [-]
The whole point of turning intelligence into a commodity is to drive its price down, not up
They are hoarding HW at massive scale, they make it harder and more expensive to own
Just because you are fine with the new price doesn't mean it's not a problem
Perhaps it's time to pop this bubble
fryanyway_swe 20 hours ago [-]
Of course. I avoid using Deepseek now that Gemini Flash is basically free on the site.
Also, Deepseek is banned in EU/US companies due to being Chinese.
During casual use Deepseek has replied to me entirely in Chinese.
Now, bring on the China glaze replies.
alkonaut 23 hours ago [-]
There is no relative/percentage increases noted (understandably). Just because i'm lazy: roughly how much more expensive is it to work with v4 flash and v4 pro through the API, compared to before the price increases? Is it 2x, 5x, 10x higher?
zupa-hu 23 hours ago [-]
# Flash, off-peak
cache-hit 2.5x
cache-miss 1.57x
out 2.36x
# Flash, peak
cache-hit 5x
cache-miss 3.14x
out 4.71x
Edit: fixed the numbers and formatting
embedding-shape 23 hours ago [-]
Someone made a comparison yesterday, including relative increases, and GPT-5.6 Luna, then later someone also added more OpenAI, Anthropic, K3 and GLM 5.2: https://news.ycombinator.com/item?id=49286679
Already outdated though I think, as GLM 5.3 is latest now :)
KronisLV 20 hours ago [-]
Oh yeah, the second one is my table (mostly just formatted what usagisushi started and added some more data points).
You'd kinda expect them to be the same as 5.2 though, seeing as that was the case with 5.1 as well (not with regular 5), but who knows.
For what it's worth, DeepSeek is still positioned as quite affordable, just not as dirt cheap as before.
wklm 19 hours ago [-]
Friendship ended with DeepSeek. Starting from Sunday gpt 5.6 luna is my best friend
floppyd 23 hours ago [-]
About 2x-2.5x off-peak for Flash, 2x-4x I'd say for Pro (x6 on cache in, the biggest increase throughout the board). And twice as much in peak hours.
23 hours ago [-]
dhx 2 days ago [-]
Old DeepSeek Flash 0731 prices have been independently reproduced.[1] The issue is DeepSeek being inundated and not having capacity to serve the demand, hence the price increases to significantly dampen demand. Never mind international demand either--just think about the magnitude of Chinese domestic demand. Prices for anything related to AI or computing in general (mobile phones, cloud data centre hosting, etc) will continue to climb fast as demand for computer chips _far_ exceeds supply. DeepSeek doesn't have an option other than to just work away on improving their technology in the period of time before computer chips once again become a commodity. For example, DeepSeek's cache ratio for their models apparently leads to 1/2 GPU time requirement versus the second best provider.[2]
Let's say DeepSeek is being forced to use the CANN stack, and the new pricing reflects the cost when 100% of inference is done with Huawei chips. Then, I suppose we can infer that:
* CANN stack is 1.5x~2.3x less efficient in compute
* CANN stack has 6x lower inter-connect capacity
> computer chips once again become a commodity
Ascend 950 is going for $7k to $9k with mediocre looking specs. $16k for RTX Pro 6000, $6k for RTX Pro 5000. This is not looking good.
dhx 2 days ago [-]
It would have to be reproduced on a CUDA stack because AFAIK Huawei don't sell the Ascend 950PR (for inference) to anyone, rather, they operate them as part of the Huawei Cloud and only allow select customers (such as DeepSeek) to rent them.
The CEO of DeepSeek recently revealed to investors a lot about the resources available to DeepSeek, and the gap between Huawei and NVIDIA. Select quotes from the transcript (translation is a bit patchy on the source website though):
"We currently have roughly 20,000 H-equivalent compute cards"
"Huawei 950—right now Huawei gives us 16,000 cards, this should be publicly stateable."
"Like Huawei gives us roughly 16,000 cards of capacity, internet giants maybe get a hundred-something thousand, we get ten-something thousand—I think this ratio is also relatively... but this is probably just how much capacity Huawei has."
"16,000 Huawei 950 cards only equal 4,000 B-series cards."
"Huawei’s supernode, Huawei’s 950 supernode, in performance and price can completely substitute for NVIDIA’s GB200, GB300. The price is definitely more expensive, but limitedly so. Fifty percent more expensive, a hundred percent more expensive—a hundred percent more doesn’t matter, two hundred percent more doesn’t matter. For example, a hundred percent more expensive—I think it can already be considered a price-level substitute."
"I think domestic hardware might need a few years."
"I don’t quite believe that five years from now, we’ll still be stuck on the production capacity problem. Right now we’re definitely stuck on the production capacity problem—this year, next year, the year after, I think we might still be stuck on the production capacity problem, but five years later, I think maybe not necessarily—I’m still relatively optimistic."
Ever since I started using flash, it has slowly crept up to be my default for everything. It is at the good enough state for a fraction of everything else that's out there.
andai 18 hours ago [-]
Have you compared it to Luna? I was using Flash for small tasks before, then switched to Luna when they dropped the price.
The benchmarks show that Luna is significantly faster, but I think those are very complex tasks for which you'd probably want a bigger model anyway. (e.g. Sol is much faster than Luna at the same tasks.)
So I'm wondering if there's any difference for smaller tasks, or if they're basically matched now.
apitman 16 hours ago [-]
The Luna price dropped the day before a massive update to Flash (0731 update). GP may be referring to that version.
alexpotato 22 hours ago [-]
I'm no expert in pricing economics but once peak/off-peak pricing arrives, it seems like tokens are going to be like electricity or long distance phone minutes where it just becomes a commodity/race to the bottom.
garrickvanburen 21 hours ago [-]
Yes. I focus on pricing software and I’m a bit baffled why frontier models are pushing tokens.
It’s a race to the bottom, and the bottom is unlimited use for a flat monthly rate.
Granular pricing (tokens, minutes, etc) is pretty anti-customer generates less revenue than customer value-based subscriptions (why SaaS is such a good business model)
progval 20 hours ago [-]
Isn't it because they have customers who will use as many tokens as they can? With a flat rate, they will run Gas Town continuously while paying as much as the occasional user.
throwup238 19 hours ago [-]
Yeah it feels like a very different model. I don’t try to fill up Apple/Google cloud drives to 1TB because then I’d have to clean up when I need space. I don’t bother trying to maximize my Audible subscription because there’s only so much I can listen to in a day. Even with my other AI subs that have monthly credits that don’t carry over, I just don’t have the interest or time to burn the credits.
But my Claude Max subscription? If I have any of my limit left the day of my reset, I’ll go and fire off research workflows with a bunch of parallel agents to explore whatever dumb ideas I had the past week. And there’s a 50:50 chance I’ll forget about it and never read the output.
butlike 18 hours ago [-]
And there's the rub. Firing off the task produces the dopamine hit, signaling you're doing something, but if you never read the output...are you really doing anything at all?
throwup238 18 hours ago [-]
I don’t know, I get dopamine hits from sharpening my handplanes and using my Veritas routers on some scrap, but I just can’t empathize with getting a dopamine hit from using a bot. Especially with how bad Opus 5 has been.
I am however going to fire off a half assed prompt when the marginal cost is zero, even if I don’t use it (which is par for the course, I probably throw out two thirds of anything the AI writes anyway be it code or prose).
Grombobulous 21 hours ago [-]
But presumably consumers aren’t where the majority of the spend will be.
Consumers don’t generally get usage-based pricing because of the inconvenience and unpredictability, but B2B SaaS products utilize usage-based pricing all the time.
dakolli 20 hours ago [-]
I've always been curious about who works on software pricing. Do you guys hire actuaries for this type of work?
kortilla 19 hours ago [-]
For software that doesn’t make sense.
Pricing software is a game of estimating both software value and the purchasing power for customers. Only the latter might have any available data and even then it won’t be sliced the right way for any in depth statistical analysis that an actuary would perform to underwrite risk.
It’s much more traditionally a more salesperson like background where being in the target market or having strong connections to it dominates efficacy.
that's a good outcome - it means they're fungible, and easily available.
vrc 21 hours ago [-]
Somehow I keep hearing the rumblings of crypto maximalists trying to merge tokens. I actually wouldn’t mind since I signed up directly with some providers I’ve stopped using and have small amounts of credits strewn across the web.
gpt-5.6-luna: $0.20 / $1.20 / $0.02 / $0.25 (In / Out / Cache Read / Cache Write)
EDIT: formatting
EDIT2: giving up on the formatting :-/
embedding-shape 2 days ago [-]
> EDIT: formatting
Keep at it, I believe in you.
KronisLV 2 days ago [-]
My apologies to any mobile users, but for the desktop folk:
Provider, Model Billing Input Output Cache read Cache write
DeepSeek
V4-Flash Old $0.1400 $0.2800 $0.0028 -
V4-Flash New Off-Peak $0.2200 (1.6x) $0.6600 (2.4x) $0.0070 (2.5x) -
V4-Flash New Peak $0.4400 (3.1x) $1.3200 (4.7x) $0.0140 (5.0x) -
V4-Pro Old $0.4350 $0.8700 $0.0036 -
V4-Pro New Off-Peak $0.6600 (1.5x) $1.9800 (2.3x) $0.0220 (6.1x) -
V4-Pro New Peak $1.3200 (3.0x) $3.9600 (4.6x) $0.0440 (12.1x) -
OpenAI
GPT-5.6 Sol $5.0000 $30.000 $0.5000 $6.2500
GPT-5.6 Terra $2.0000 $12.000 $0.2000 $2.5000
GPT-5.6 Luna $0.2000 $1.2000 $0.0200 $0.2500
Anthropic
Claude Fable 5 $10.000 $50.000 $1.0000 $12.500
Claude Opus 5 $5.0000 $25.000 $0.5000 $6.2500
Claude Sonnet 5 $2.0000 $10.000 $0.2000 $2.5000
Moonshot
Kimi K3 $3.0000 $15.000 $0.3000 -
Z.AI
GLM 5.2 $1.4000 $4.4000 $0.2600 -
worik 13 hours ago [-]
Forgive my naivety... What are the units? $1.4000 for how many input tokens?
smrtinsert 1 days ago [-]
Seems to still be cheaper at worst case scenario on a pay as you schedule. The price increase isn't ideal, but still seems like a good deal to me.
jwrallie 1 days ago [-]
There are many nuances like amount of cache hit, the time of the day you use it and how other providers or even competitors respond, so it is hard to say concretely, but it might change my monthly usage close to one of the $20 USD plans.
I’m curious about how openrouter and Luna prices will change in response.
KronisLV 1 days ago [-]
Yeah, seems the same for me! Except the subscription plans you can get like Kimi K3 or GLM Coding Plan are still giving you more value if you need that many tokens (and can use those plans).
usagisushi 2 days ago [-]
TIL: formatting tables on HN is computationally impossible. lol
p.s. Thanks DSv4-Flash, for your hard work of converting a messy table into plain text.
roenxi 22 hours ago [-]
This is somewhat funny when you realise the data centres are now going to start a process that looks very so slightly like daydreaming. Depending on the time of day they're going to be thinking about different things in a cyclic manner. They're going to be doing things like finishing a hard days work then kicking back to think about tricky math problems.
squidbeak 20 hours ago [-]
It's worth keeping in mind the model doesn't keep a running memory. Each time its instantiated, it begins from its release state - so from its perspective (if it had one) the current task would be the first stop after posttraining. Perhaps the only stop.
Though of course you're talking about data centers, and romanticizing them rather than the AI itself.
halJordan 19 hours ago [-]
No, llm providers will start providing a service that looks a lot like rumination or (day)dreaming. Like thats the prompt "you're daydreaming about this work you recently did" then add in whatever is in the current session.
butlike 18 hours ago [-]
Why do they do it this way? Because dogfooding is harmful to the model? Is this implying there's no benefit in having the model train on itself?
HawtAds 17 hours ago [-]
> Is this implying there's no benefit in having the model train on itself?
There are various theories around model collapse when you train on too much AI generated data (that's not for distillation).
Grombobulous 21 hours ago [-]
That’s an interesting thing to think about. Still, it’s important for us to remind ourselves that “looks very slightly like” is not the same as the real thing. The A in AI stands for artificial.
The summary of this paper describes my sentiment in better words than I have:
It’s very easy for the average person to mistake linguistic ability and simulated problem solving for intelligence and sentience.
s08148692 20 hours ago [-]
Without knowing what makes consciousness possible, the paper cannot justify biology as necessary - it mistakes a lack of evidence for conscious AI for proof that conscious AI is impossible.
I do not believe current AI or LLMs are conscious, but there is no proof one way or another that they can or cannot be. The paper authors are making up their own definitions and building an argument from them
Grombobulous 12 hours ago [-]
I would make one further argument: as the top-of-the-food-chain species which originated the concept of consciousness, we get to define it however we want.
Consciousness is defined in the human context. We can just say “hey buddy that’s not biological enough to qualify.”
blazarquasar 19 hours ago [-]
Your argument requires that there is some objective truth for what consciousness is. It will always hinge on what definition one accepts.
I, and apparently many others, don’t think it would be any useful to describe the mathematical properties of an AI as consciousness. To me it is inherently a way to describe the “experience” arising from physical processes in biological beings as ourselves.
That’s what the argument comes down to for me. Could an LLM “fall unconscious”?
halJordan 19 hours ago [-]
I think you, and apparently many others, are hiding behind mathematical strictures to avoid the discussion. Your argument is actually "what requires an objective truth" Or else you could admit there's no real difference between zapping amino acids with electricity and zapping silicon with electricity.
sinsterizme 18 hours ago [-]
Why are AIs mathematical beings with no possibility of consciousness whereas humans fall outside this and have the capability of consciousness?
HarHarVeryFunny 21 hours ago [-]
We'll have Dwarkesh's "datacenter full of geniuses" with 99% of the geniuses coding up CRUD apps, then the dusty GPU in the corner, with the "do not disturb" sign on it, pipes up "You're absolutely right! The answer is 42!".
ssk42 22 hours ago [-]
That’s what my KimiClaw has literally been doing
petercooper 2 days ago [-]
As well as the headline in/out changes, people heavily using agentic coding tools will want to note the 6x (off peak) and 12x (peak) increase to cache hit pricing on Pro (since cache hit can easily make up 90%+ of input on long sessions).
DeepSeek was hugely underpricing cache hit pricing before and even after this increase they're still cheaper on that metric than every other provider I'm aware of, but it will put an end to those "I used 1 billion tokens and spent $4" reports.
benjiro29 2 days ago [-]
The problem with DS Flash/Pro is that they are extreme reasoning heavy and step heavy. Step = cache hit. Reasoning = output hit. So the impact on those price increases will be felt much stronger.
I think that Flash is still a usable model but Pro is DOA... Even before the price difference between Flash and Pro, vs the intelligence / problem solving / tool calling did not make sense. But now that gap has widen even more. And there are just too many competitors models now close to that Pro price range.
Especially when we compare that competitive models offer subscription services that easily cut down the token price by 1:10. That makes Pro especially a bad value.
We shall see what the 3th party market is going to do, but i suspect that prices will be increased. If the argument was that DeepSeek increases price as they lack capacity, a company with access to billions, other 3th party providers that need to rent and have less optimized infrastructures will increase prices. Especially if they get hit hard with people moving around.
Its like we always see the same issue with popular models.
* GLM 5.2 is good, capacity issues, API price up, subscription heavy nerfs.
* Kimi K3 is good, capacity issues, API price up, subscription heavy nerfs.
* DeepSeek V4 GA is good, capacity issues, API price up
* OpenAI GLM 5m, 10m active users. Subscription usage is sneakily tightened more and more.
* Anthropic Opus too popular, ...
That is the main issue. The AI users are people who actively easily move between companies. Pushing peak loads to each unprepared company, releasing load on the "less desired". And round we go ...
slopinthebag 1 days ago [-]
I haven’t noticed the Deepseek models being especially verbose. They’re also so cheap to run it doesn’t matter. These pricing changes are inconsequential since even 100 * ~0 is still a low number.
andai 2 days ago [-]
So OpenAI cut Luna's price by 5x, DeepSeek increased price by 5x!
If I'm reading the benchmarks right, they now went from being much cheaper than Luna (but twice as slow), to being roughly same price (but twice as slow).
So all else being equal, where I would previously have used DeepSeek, I can just use Luna, and get the same result twice as fast?
(Yeah I know benchmarks are mostly nonsense, but the ones measuring time are real, and it's the most precious resource.)
culi 2 days ago [-]
DeepSeek prices only increased by 3-5x if you look at peak pricing only (In California that's 6-9pm and 11pm-3am. So not even the typical workday).
Also if you're considering Luna, I assume you don't care about this but I think it's worth pointing out: a major advantage of DS is the ability to self-host or choose a different host. As a customer that gives you much more negotiating power and potential privacy guarantees.
I don’t trust this bench. I move between opus and ds v4 pro and find them both significantly better then Gemini 3.7
hopfenspergerj 22 hours ago [-]
Does the API response include a "service tier" response to indicate whether you paid peak/off-peak for a given request? I like to compute cost for each request, and save it with my results.
f311a 2 days ago [-]
Opencode said they are working on matching the old prices using their own inference.
Right now, they give 4100 credits for Luna and 63 000 for Deepseek on their prepaid plan (both are 2x)
Tharre 2 days ago [-]
How could that possibly work? Deepseek was undercutting every other provider by an order of magnitude on cached tokens.
Do they just set a super low caching time and hope that drops effective cache rates low enough? Do all other providers somehow overcharge by that much? Are they just going to sell it as a loss leader?
lcampbell 2 days ago [-]
> Do all other providers somehow overcharge by that much?
This, I think. Cached inputs have an opportunity cost (keeping the KV cache until use) but a hit is basically free. “Basically” - if the cache is offloaded to system RAM or NVMe there’s some scheduling overhead.
From a consumer viewpoint a more interesting metric than the raw costs is
from a personal standpoint rather than a per-provider one (e.g. if OpenRouter is blindly dispatching your requests you might have a bad time).
skeledrew 2 days ago [-]
> if OpenRouter is blindly dispatching your requests
This can somewhat be the case, depending on your config. I updated mine to make DeepSeek high priority because I was having a lot of cache misses and reliability issues with the default (cheapest (at face value)) providers, and cost was actually higher overall than anticipated. Was smooth sailing from then; might have to tweak things again now pricing has changed though.
nyargh 2 days ago [-]
I was having the same issue. Did not find any provider that had a cache hit rate anywhere close to DeepSeeks own API.
Not sure if this was an OpenRouter issue or with the other inference providers.
brandon997 1 days ago [-]
You could also check out token.dance. They offer DeepSeek and other models through an OpenAI-compatible API, and the pricing seems pretty competitive. Might be worth comparing the cache rates.
Tharre 2 days ago [-]
System RAM and/or NVMe storage still has a real cost. And swapping out the context between VRAM and system RAM / NVMe still consumes bandwidth.
I don't have a clue on what the real cost to inference providers comes out to, but it seems really weird that there would be such a big gap, in what should be a pretty competitive market.
2 days ago [-]
pzo 2 days ago [-]
I doubt they will match old cache read pricing- that’s most important in agentic coding.
jauntywundrkind 2 days ago [-]
CXL might save us. All the world's old DDR4 to the rescue. Either per box, where the job has to route back, or network attached, where there's now a pool of absurdly fast temp storage.
j1elo 21 hours ago [-]
So many changes in so little time, that it all makes no sense. Continuous churning. Reminds me of the experience of trying to be on top of the dependencies in a medium-large JS project.
I am a person that buys into a tool or a process and expects it to be part of the life with no major changes through the years (or as long as the need exists). But AI? You buy into something today, not 2 weeks have passed and there's already a large "update" introduced to the conditions or the optimal usage patterns you should be adopting.
It's tiring. Makes all prices and offers feel so unreliable and gets me a bit more disinterested each time they change.
vikramkr 18 hours ago [-]
Always gonna exist during a period of rapid exploration and experimentation. The js/web dev world slowed down a lot and entered a steady state eventually - been years since react took over and nothing's displaced it since
ricardobeat 21 hours ago [-]
This makes no sense. You want improvements to stop?
These being open, you can keep using the old models indefinitely for as long as there are providers offering them.
j1elo 21 hours ago [-]
No, I'm talking about the whole sector, not specifically about DeepSeek.
Fully knowing that it is a new industry living its own infancy, it is perfectly normal that there is instability and numerous swings on pricing, conditions, or direction.
But it's not less real that such process can produce churn and consumer fatigue.
ChildOfChaos 2 days ago [-]
I pay for Google AI Pro (Bought a year in advance) and Gemini is so bad, I burned through 75% of my five hour allowance trying to get it to fix something.
I pasted the same prompt into OpenCode, set to Deepseek v4 flash free and did it first try.
I'm was going to purchase Opencode GO to try it, but seems my timing is really bad :( hope it doesn't go up too much in Opencode or they find other providers. Bad timing!
brandon997 1 days ago [-]
If you’re worried about pricing, token.dance might be worth a look. It offers multiple AI models through one OpenAI-compatible API at competitive prices.
scld 2 days ago [-]
I just bought a month of GO two days ago, I feel you. However, given access to both Flash and Luna there, probably still worth the $10 a month. I've been very happy with Flash's ability to implement a detailed, focused plan from Sol/Opus.
unified101 2 days ago [-]
About 3x increase. Luna is now a much better deal. Hope they don't increase their prices in response.
eli 2 days ago [-]
Luna has one vendor and they can change the price any time. Deepseek is open and has dozens of vendors competing to serve it.
drillsteps5 2 days ago [-]
Can you help me find a few names? These are inference providers running open weights models, right?
Thank you in advance!
brandon997 1 days ago [-]
You might want to check out token.dance. It offers multiple AI models through one OpenAI-compatible API, with very competitive pricing.
apitman 2 days ago [-]
Look at OpenRouter. They have lots of providers
cmrdporcupine 2 days ago [-]
Luna vs v4 Flash, sure.
But DeepSeek v4 Pro is a far more capable model and still cheaper than anything that it competes with, from what I can see.
procgen 2 days ago [-]
Luna is multimodal.
cmrdporcupine 2 days ago [-]
Not sure who you're replying to? I was talking about Pro not Flash. Luna and DS4Pro are not competitors. DS4 Pro is probably about GPT 5.4 high level or Opus 4.8, not Luna. It's cheaper than either.
Flavius 2 days ago [-]
Why not? They probably only offered those deals because of DeepSeek aggressive pricing. Now that DeepSeek is 3x more expensive it's time to revert those discounts.
m101 2 days ago [-]
Tibo said it’s permanent, whatever that means
LoganDark 2 days ago [-]
Probably means it's not intended as temporary, but not that the price will never change again
declan_roberts 18 hours ago [-]
This actually works out favorably for US customers since the peak hours are Chinese working hours and cheap hours are US working hours.
PeterStuer 17 hours ago [-]
Not yet, but in the end high quality tokens are a commodity market. Every optimization to increase inference efficiency will be universally rolled out. The 'hyperspenders' will run into demishing returns unless regulatory capture succeeds.
poly2it 23 hours ago [-]
That's a hefty increase. Flash pricing during peak is now 1.32/M out, compared to the current 0.28/M, which in turn is a quite a bit above the cheapest provider at 0.16/M.
You don't mention the off-peak pricing which is half that. Most western workplaces will enjoy the cheaper prices.
poly2it 18 hours ago [-]
Yes, I explicitly said during peak. It's the data point I found most interesting, as it's almost an order of magnitude more expensive than their competitors.
HarHarVeryFunny 21 hours ago [-]
Some of the US companies do the same, but rather than "off-peak" hours they price lower for "batch" jobs with non-committal response times.
The same motivation of course - the GPUs have a finite service lifetime, so to maximize revenue you need to keep them busy 24x7.
sebastiennight 22 hours ago [-]
With proprietary labs lowering their prices and Deepseek raising theirs over time, wouldn't it possible to extrapolate a graph to look at where the terminal frontier-model million-token-cost asymptotes to?
dakolli 20 hours ago [-]
off of one historical price change, no.
xbmcuser 21 hours ago [-]
They benefit from a strong captive market because Chinese firms cannot use Nvidia chips and are legally barred from processing data abroad, forcing them to rely on domestic infrastructure.
snake_doc 21 hours ago [-]
lol what no, AI competition is super competitive in China; bytedance has 50% of the inference market and mostly serves from outside of China
Bytedance which runs China’s most popular Doubao AI chatbot; is spending $70B in CapEx this year, most of it outside of China (Malaysia, Thailand, Brazil, etc; and they are allowed to lease NVIDIA chips). This is roughly 50% of Microsoft CapEx.
This is good for other competitors I guess. People rarely calculate the bump in price but the fact that price is increasing might bring them to other vendors.
kubafu 19 hours ago [-]
Are we gonna see "we work those unusual hours because that's when LLMs are cheap"?
cheesecakegood 23 hours ago [-]
I wonder if this is enough to push people back onto Luna with their comparative price drop
eastbound 23 hours ago [-]
After the big onshore migrations (startup people migrating to the SV),
The big Covid migrations (startup prople migrating to the countryside),
Will we see the big AI migrations (people travelling to where AI is the cheapest)?
reddec 2 days ago [-]
It doesn't make any sense unless they are going to exit from inference market. They will be literally one of the costliest option (by output, for flash) if use openrouter as source.
20 hours ago [-]
kortzeus 2 days ago [-]
Makes sense, basically increased price for peak hours when they don't have enough infra to serve everyone. Can expect the return of prices back.
pu_pe 2 days ago [-]
This now places deepseek flash v4 from DeepSeek themselves at higher prices than openrouter (depending on caching). Will be interesting to see if third party prices remain the same.
cmrdporcupine 2 days ago [-]
It does seem to me that DeepSeek themselves aren't so much interested in being a service provider. They do it, and offer the service, but their pronouncements seem to be that they're more interested in being for now closer to a research lab with a longer term play for something more dramatic later.
flakiness 19 hours ago [-]
Now Baseten's pricing is cheaper than the official one? Probably won't last, but still interesting.
the neat thing about the models being open-weights is there's a dozen other providers on OpenRouter still at the old price, or lower
Tiberium 2 days ago [-]
Unfortunately in this case that's not true at all, there was no provider with genuinely close or the same effective prices (mostly based on cache hit cost) to old v4 flash or v4 pro. People have this misconception that other providers must be much cheaper than the official one in case of open weight models.
If you check on OpenRouter, some other providers serve V4 Flash at seemingly cheaper normal input/output tokens rates, but with a huge caveat: they have at least a 5x increase of the cache hit cost of the official API, some have a 10x+. No provider comes close to Deepseek's old low cache prices, and cache is 90%+ of what matters in agentic sessions.
Closest comparison:
- Deepseek: $0.14/$0.28 with $0.0028 cache hit cost for official API
- DeepInfra: $0.08/$0.18 (cheaper base rates!) with $0.016 cache hit (almost 6x!! Deepseek's current cache cost)
Another great example is Kimi K3, official API is $3/$15 and the cheapest provider on OpenRouter is $2.8/$14, only a tiny difference.
quikoa 2 days ago [-]
The question is if they'll increase prices as well.
slopinthebag 2 days ago [-]
Ah well. I spend about $5/month with Deepseek, so now I’ll have to find room in my budget for $15. Might have to tip my barista less or something.
alexthedigger 20 hours ago [-]
Now if there could be a bot that defers queries until when it’s cheap…
Sha1rholder 15 hours ago [-]
"half-priced batch API" kind of thing has been around for a long time.
skeledrew 20 hours ago [-]
Wouldn't be hard to whip up...
butlike 18 hours ago [-]
The whipping's the easy part. It's the plating and serving that takes time
benniemack 2 hours ago [-]
[flagged]
floppyd 23 hours ago [-]
Full table with multipliers from previous prices:
DeepSeek-V4-Flash (off-peak, x2 for peak)
* Cache Hit $0.007 (x2.5)
* Cache Miss $0.22 (x1.5)
* Output $0.66 (x2.25)
DeepSeek-V4-Pro (off-peak, x2 for peak)
* Cache Hit $0.022 (x6)
* Cache Miss $0.66 (x1.5)
* Output $1.98 (x2.25)
Peak Hours: 01:00–04:00 and 06:00–10:00 UTC
Effective from: 16:00, August 16, 2026 (UTC)
spuz 23 hours ago [-]
I wonder whether all the DeepSeek providers will follow suit or are they going to try to stay competitive with the old prices?
trollbridge 22 hours ago [-]
DeepSeek’s cache pricing was always 1/10th the competition.
It’s still cheaper than everybody else.
k__ 22 hours ago [-]
I didn't get the impression that anyone competed with the old prices before.
spuz 22 hours ago [-]
What do you mean? Most providers on OpenRouter offer the same or lower prices than DeepSeek themselves:
Anthropic doesn't allow Claude in China and neither does OpenAI. But the rest of Asia certainly has a choice. Unless you mean socioeconomically. Also there's plenty of loopholes people in China can use to still access those models
If you buy Claude api access through a third party Chinese company, it's cheaper than when you buy directly.
Beside it would go against the Chinese philosophy of just using the right tool for the job. If Claude is better at a task than deepseek, they sure use Claude.
No really I don't get your claim, do you have any proof that you can source ??
Deepseek's official API has a pretty bad privacy policy so I would assume businesses avoid them in any event
Interestingly it seems that Chinese customers are even more privacy-concerned than US ones, which is why the majority of Ziphu's (GLM) business is support services to Chinese companies running their open-weight models on-prem!
Not saying they are or will, but their privacy policy is so permissive.
Which is a real bummer because it’s otherwise solid with excellent caching.
I'd still prefer it wasn't like that, but I guess it's a compromise ultimately, at least I'll be able to run it myself.
I know it's more complicated than that, with economics and privacy factors involved. But if I operate on the assumption that real privacy is real hard I'd rather just be guarded and careful with my prompts and whatever output I give to the LLM and expect that it will be training on that, rather than spill all my darkest secrets to some other LLM provider that pinkie-promises privacy only to leak it publicly later, anyway.
That sounds like a policy written by someone who doesn't understand how LLM's work...
Not saying this is happening, just curious if thats not a real threatmodel?
It’d be much easier to hide sketchy code in an agent harness, but “vendor adds spyware to their software” isn’t a novel issue.
I think the only sort of new issue is people “allow all”ing their agents tool calls, but that’s more or less the same issue as curl | bash
Pi/OpenCode seem pretty straight foward and widely used enough for this to be viable
OMP as I understand does it's own vendoring of tools, so I assume it'd be a pain in the ass to audit, but that means you're even safe from base OS shenanigans
I imagine it would be very non trivial to do it in a way that that was reliable and obfuscated enough to prevent detection for any amount of time?
Hopefully this will change soon. But AI and China/US skepticism is very high. Even if the person you talk to isn't skeptic, his boss may be. And even if his boss isn't, his CFO or Legal department may use it as a political lever and therefore if you can say 'everything in europe' you dodge the tension entirely.
Yeah it's dumb.
Why use a Chinese product when a domestic or EU one is better and safer?
For European and US customers this is effectively 2x increase. I think i wll keep using both Flash and Pro as before.
EDIT: Misread numbers to believe off-peak kept old prices
Personally, I don't think we've seen the total end of dirt cheap LLMs, it's just a frontier lab doesn't want to be in business of serving half the world.
You certainly don't need Fable to code up a basic web app, any more than you need a Ferrari to go grocery shopping.
As someone from just such a country, DeepSeek 0731 was the first time I seriously started using an LLM for coding. All previous attempts were useless or ridiculously expensive.
Can't say the old prices felt "free", but it was affordable if you're careful with your cache hit rate.
The new pricing probably pushed it into the unaffordable territory for tasks where you can do without it. Probably will try opencode go if they don't also follow suite, or will have to go back to wetware.
They are hoarding HW at massive scale, they make it harder and more expensive to own
Just because you are fine with the new price doesn't mean it's not a problem
Perhaps it's time to pop this bubble
Also, Deepseek is banned in EU/US companies due to being Chinese.
During casual use Deepseek has replied to me entirely in Chinese.
Now, bring on the China glaze replies.
Already outdated though I think, as GLM 5.3 is latest now :)
I am already using GLM 5.3 with their coding plan, but oddly enough the API prices don't really seem to be out yet: https://docs.z.ai/guides/overview/pricing
You'd kinda expect them to be the same as 5.2 though, seeing as that was the case with 5.1 as well (not with regular 5), but who knows.
For what it's worth, DeepSeek is still positioned as quite affordable, just not as dirt cheap as before.
[1] https://nitter.net/thdxr/status/2085377844515922210#m
[2] https://nitter.net/thdxr/status/2087610161636471289#m
Let's say DeepSeek is being forced to use the CANN stack, and the new pricing reflects the cost when 100% of inference is done with Huawei chips. Then, I suppose we can infer that:
* CANN stack is 1.5x~2.3x less efficient in compute
* CANN stack has 6x lower inter-connect capacity
> computer chips once again become a commodity
Ascend 950 is going for $7k to $9k with mediocre looking specs. $16k for RTX Pro 6000, $6k for RTX Pro 5000. This is not looking good.
The CEO of DeepSeek recently revealed to investors a lot about the resources available to DeepSeek, and the gap between Huawei and NVIDIA. Select quotes from the transcript (translation is a bit patchy on the source website though):
"We currently have roughly 20,000 H-equivalent compute cards"
"Huawei 950—right now Huawei gives us 16,000 cards, this should be publicly stateable."
"Like Huawei gives us roughly 16,000 cards of capacity, internet giants maybe get a hundred-something thousand, we get ten-something thousand—I think this ratio is also relatively... but this is probably just how much capacity Huawei has."
"16,000 Huawei 950 cards only equal 4,000 B-series cards."
"Huawei’s supernode, Huawei’s 950 supernode, in performance and price can completely substitute for NVIDIA’s GB200, GB300. The price is definitely more expensive, but limitedly so. Fifty percent more expensive, a hundred percent more expensive—a hundred percent more doesn’t matter, two hundred percent more doesn’t matter. For example, a hundred percent more expensive—I think it can already be considered a price-level substitute."
"I think domestic hardware might need a few years."
"I don’t quite believe that five years from now, we’ll still be stuck on the production capacity problem. Right now we’re definitely stuck on the production capacity problem—this year, next year, the year after, I think we might still be stuck on the production capacity problem, but five years later, I think maybe not necessarily—I’m still relatively optimistic."
[1] https://www.fredgao.com/p/deepseeks-liang-wenfeng-breaks-his
The benchmarks show that Luna is significantly faster, but I think those are very complex tasks for which you'd probably want a bigger model anyway. (e.g. Sol is much faster than Luna at the same tasks.)
So I'm wondering if there's any difference for smaller tasks, or if they're basically matched now.
It’s a race to the bottom, and the bottom is unlimited use for a flat monthly rate.
Granular pricing (tokens, minutes, etc) is pretty anti-customer generates less revenue than customer value-based subscriptions (why SaaS is such a good business model)
But my Claude Max subscription? If I have any of my limit left the day of my reset, I’ll go and fire off research workflows with a bunch of parallel agents to explore whatever dumb ideas I had the past week. And there’s a 50:50 chance I’ll forget about it and never read the output.
I am however going to fire off a half assed prompt when the marginal cost is zero, even if I don’t use it (which is par for the course, I probably throw out two thirds of anything the AI writes anyway be it code or prose).
Consumers don’t generally get usage-based pricing because of the inconvenience and unpredictability, but B2B SaaS products utilize usage-based pricing all the time.
Pricing software is a game of estimating both software value and the purchasing power for customers. Only the latter might have any available data and even then it won’t be sliced the right way for any in depth statistical analysis that an actuary would perform to underwrite risk.
It’s much more traditionally a more salesperson like background where being in the target market or having strong connections to it dominates efficacy.
that's a good outcome - it means they're fungible, and easily available.
EDIT: formatting
EDIT2: giving up on the formatting :-/
Keep at it, I believe in you.
I’m curious about how openrouter and Luna prices will change in response.
p.s. Thanks DSv4-Flash, for your hard work of converting a messy table into plain text.
Though of course you're talking about data centers, and romanticizing them rather than the AI itself.
There are various theories around model collapse when you train on too much AI generated data (that's not for distillation).
The summary of this paper describes my sentiment in better words than I have:
https://www.nature.com/articles/s41599-025-05868-8
It’s very easy for the average person to mistake linguistic ability and simulated problem solving for intelligence and sentience.
I do not believe current AI or LLMs are conscious, but there is no proof one way or another that they can or cannot be. The paper authors are making up their own definitions and building an argument from them
Consciousness is defined in the human context. We can just say “hey buddy that’s not biological enough to qualify.”
I, and apparently many others, don’t think it would be any useful to describe the mathematical properties of an AI as consciousness. To me it is inherently a way to describe the “experience” arising from physical processes in biological beings as ourselves.
That’s what the argument comes down to for me. Could an LLM “fall unconscious”?
DeepSeek was hugely underpricing cache hit pricing before and even after this increase they're still cheaper on that metric than every other provider I'm aware of, but it will put an end to those "I used 1 billion tokens and spent $4" reports.
I think that Flash is still a usable model but Pro is DOA... Even before the price difference between Flash and Pro, vs the intelligence / problem solving / tool calling did not make sense. But now that gap has widen even more. And there are just too many competitors models now close to that Pro price range.
Especially when we compare that competitive models offer subscription services that easily cut down the token price by 1:10. That makes Pro especially a bad value.
We shall see what the 3th party market is going to do, but i suspect that prices will be increased. If the argument was that DeepSeek increases price as they lack capacity, a company with access to billions, other 3th party providers that need to rent and have less optimized infrastructures will increase prices. Especially if they get hit hard with people moving around.
Its like we always see the same issue with popular models.
* GLM 5.2 is good, capacity issues, API price up, subscription heavy nerfs. * Kimi K3 is good, capacity issues, API price up, subscription heavy nerfs. * DeepSeek V4 GA is good, capacity issues, API price up * OpenAI GLM 5m, 10m active users. Subscription usage is sneakily tightened more and more. * Anthropic Opus too popular, ...
That is the main issue. The AI users are people who actively easily move between companies. Pushing peak loads to each unprepared company, releasing load on the "less desired". And round we go ...
If I'm reading the benchmarks right, they now went from being much cheaper than Luna (but twice as slow), to being roughly same price (but twice as slow).
So all else being equal, where I would previously have used DeepSeek, I can just use Luna, and get the same result twice as fast?
(Yeah I know benchmarks are mostly nonsense, but the ones measuring time are real, and it's the most precious resource.)
Also if you're considering Luna, I assume you don't care about this but I think it's worth pointing out: a major advantage of DS is the ability to self-host or choose a different host. As a customer that gives you much more negotiating power and potential privacy guarantees.
Right now, they give 4100 credits for Luna and 63 000 for Deepseek on their prepaid plan (both are 2x)
Do they just set a super low caching time and hope that drops effective cache rates low enough? Do all other providers somehow overcharge by that much? Are they just going to sell it as a loss leader?
This, I think. Cached inputs have an opportunity cost (keeping the KV cache until use) but a hit is basically free. “Basically” - if the cache is offloaded to system RAM or NVMe there’s some scheduling overhead.
From a consumer viewpoint a more interesting metric than the raw costs is
from a personal standpoint rather than a per-provider one (e.g. if OpenRouter is blindly dispatching your requests you might have a bad time).This can somewhat be the case, depending on your config. I updated mine to make DeepSeek high priority because I was having a lot of cache misses and reliability issues with the default (cheapest (at face value)) providers, and cost was actually higher overall than anticipated. Was smooth sailing from then; might have to tweak things again now pricing has changed though.
Not sure if this was an OpenRouter issue or with the other inference providers.
I don't have a clue on what the real cost to inference providers comes out to, but it seems really weird that there would be such a big gap, in what should be a pretty competitive market.
I am a person that buys into a tool or a process and expects it to be part of the life with no major changes through the years (or as long as the need exists). But AI? You buy into something today, not 2 weeks have passed and there's already a large "update" introduced to the conditions or the optimal usage patterns you should be adopting.
It's tiring. Makes all prices and offers feel so unreliable and gets me a bit more disinterested each time they change.
These being open, you can keep using the old models indefinitely for as long as there are providers offering them.
Fully knowing that it is a new industry living its own infancy, it is perfectly normal that there is instability and numerous swings on pricing, conditions, or direction.
But it's not less real that such process can produce churn and consumer fatigue.
I pasted the same prompt into OpenCode, set to Deepseek v4 flash free and did it first try.
I'm was going to purchase Opencode GO to try it, but seems my timing is really bad :( hope it doesn't go up too much in Opencode or they find other providers. Bad timing!
Thank you in advance!
But DeepSeek v4 Pro is a far more capable model and still cheaper than anything that it competes with, from what I can see.
https://openrouter.ai/deepseek/deepseek-v4-flash-0731#provid...
The same motivation of course - the GPUs have a finite service lifetime, so to maximize revenue you need to keep them busy 24x7.
https://www.bloomberg.com/news/articles/2026-06-17/microsoft...
Bytedance which runs China’s most popular Doubao AI chatbot; is spending $70B in CapEx this year, most of it outside of China (Malaysia, Thailand, Brazil, etc; and they are allowed to lease NVIDIA chips). This is roughly 50% of Microsoft CapEx.
https://www.tomshardware.com/pc-components/gpus/chinas-byted...
The big Covid migrations (startup prople migrating to the countryside),
Will we see the big AI migrations (people travelling to where AI is the cheapest)?
https://www.baseten.co/pricing/
If anyone has tried Baseten versions of these Chinese frontier models, let me know what you found.
If you check on OpenRouter, some other providers serve V4 Flash at seemingly cheaper normal input/output tokens rates, but with a huge caveat: they have at least a 5x increase of the cache hit cost of the official API, some have a 10x+. No provider comes close to Deepseek's old low cache prices, and cache is 90%+ of what matters in agentic sessions.
Closest comparison:
- Deepseek: $0.14/$0.28 with $0.0028 cache hit cost for official API
- DeepInfra: $0.08/$0.18 (cheaper base rates!) with $0.016 cache hit (almost 6x!! Deepseek's current cache cost)
Another great example is Kimi K3, official API is $3/$15 and the cheapest provider on OpenRouter is $2.8/$14, only a tiny difference.
DeepSeek-V4-Flash (off-peak, x2 for peak)
* Cache Hit $0.007 (x2.5)
* Cache Miss $0.22 (x1.5)
* Output $0.66 (x2.25)
DeepSeek-V4-Pro (off-peak, x2 for peak)
* Cache Hit $0.022 (x6)
* Cache Miss $0.66 (x1.5)
* Output $1.98 (x2.25)
Peak Hours: 01:00–04:00 and 06:00–10:00 UTC
Effective from: 16:00, August 16, 2026 (UTC)
It’s still cheaper than everybody else.
https://openrouter.ai/deepseek/deepseek-v4-flash#providers
- but what matter - is cache hit
even now deepseek's off-peak hours for cache hit (0.007) is lower than other providers (~0.01)