Posted by jonotime 21 hours ago
I tried the cheapest provider on openrouter and burned through $50 in a few days. Quality was ok, seems slightly above Luna quality perhaps? But that $50 is 1/4 of my codex subscription where I could have burned that many tokens or more using Astra within my weekly reset.
This won’t last forever but as long as the frontier labs are subsidizing this heavily the open models won’t matter.
I am having 98% my input in cache, so using Coralbricks makes sense due to them giving cache reads for free — you only pay for writes. I spend maybe 5-10 dollars a day and my agents basically work day and night implementing things for me.
If your tasks are write-heavy, find a provider with cheaper output.
If you build a customer-facing app, pay a bit extra for 400+ tok/s e.g. on Lithos.
And I am using the Claude Code harness with DS as the endpoint. And I use it ~5-8hrs a day to do my coding.
Buy directly from DeepSeek's API.
You can literally get overcharged 100x on DeepSeek on OpenRouter (or more).
OpenRouter Pricing:
$0.02/M input tokens $0.60/M output tokens
DeepSeek Pricing (cache miss, off-peak):
$0.15/M Input $0.60/m output
It’s also the major difference between using DeepSeek directly vs other providers also serving it, though I have not looked lately: it’s possible other providers have matched its cache hit pricing better?
VRAM & Memory Requirements by Precision
• FP16 (Full Precision): Requires ~1,664 GB of VRAM (e.g., an 8x B300 288GB cluster).
• INT8 Quantization: Requires ~832 GB of VRAM (e.g., 8x H200 141GB).
• INT4 Quantization: Requires ~416 GB of VRAM (e.g., 8x A100 80GB)
VRAM aint cheap, Sam Altman ruined the cost of memory, Nvidia doesnt make enough consumer GPUs letting the market go insane over them, I still have friends on 1070s or 1070 TIs because GPUs have been severely overpriced for too long. I remember when a gaming PC was only $1000.
Even so why would anyone not sleep on a model they cannot run?
Memory companies have price fixed multiple times. They've paid hundreds of millions in fines. wikipedia even has a page on it. https://en.wikipedia.org/wiki/DRAM_industry_price_fixing.
Look at the financials of these companies, they're all making obscene margins and do they plan to increase production? No. Micron is doing a stock buy back to pump the price of their share.
The Micron CEO just recently said this is the exact plan https://www.theregister.com/systems/2026/10/01/ram-supply-se...
There's sanctions, tarrifs, and a DOJ who doesn't give a shit. Until we can fix that the insanity will continue. Phones will be unaffordable. Laptops will be obscene. Gaming consoles will be thousands of dollars. Desktops will be dead.
If you're waiting for some David Ricardo equation to happen, tough cookies, it's not coming.
The market is legally locked down and we're in hostage pricing mode.
And what's the story? You can't afford electronics because we're using it to build robots to take your job? I mean ...
Nobody is coming to save us. That's our job.
Micron has 3 brand new fabs currently under construction, 2 Boise, 1 in New York as the first of 4 planned for a campus.
Plus expanding other existing facilities.
These things take ~3-5 years from breaking ground to full production. You'd have had to anticipate the current demand years before it happened in order to be bringing production on-line before 2030 or so.
Samsung and HK Hynix also have fabs under construction and planned.
CXMT started 11 years ago and only now is reaching any real volume. If they decided a year ago to react to the current demand cycle they'd be 6-7 years out.
Not much you can really do to wish for more fabrication to exist on any timeline not measured in fractional decades.
Could they do more and react quicker? Probably, but everything I've read on the subject seems to point to 3 years is absolute bare minimum if you happen to have a shovel ready project with the land bought, local permitting completed, infrastructure extended to the site, and a skilled workforce already in place. They could suspend buy-backs/dividends today and dump it all into building production and there would be no material impact until around 2030.
> The Micron CEO just recently said this is the exact plan
CEO simply stated the demand pressure will not go away through 2027, and supply will not increase until around 2028 when currently under construction fabs start shipping volume. The article does not support your statement.
Costs did go nuts, but there are signs of easing in the market of late. CXMT is starting to have an impact and priced will probably fall in 2027.
Some of my family is pretty happy, though, with the job security as they are pretty convinced these projects are all going to take much longer than what's being stated publicly. Micron is saying the first chip from the new fab will be in 2027... though they also predicted it'd be 2026. The date seems pretty slippy.
You need to keep the market healthy, not some insane Bitcoin style HODL pump - that's how you get wrecked.
I mean I'm not a neoclassicalist but I've read all of them. I'm in consensus with them here. There's a bunch of theories on what a healthy market is but what we're currently seeing matches none of them.
It's short term profitable but long term disastrous, especially in a world where new mathematics and techniques could literally collapse the demand overnight.
Imagine if some paper hits arxiv and the 256 GB requirement for some model now becomes 64. Woops!
Some clever trick about how attention heads and context Windows work could potentially slash a bunch of requirements by giant margins and all they're doing is firing the starting gun at that global race with every obscenely priced unit they sell.
But if prices were reasonable, this wouldn't be an apocalypse. It'd be fine. Consumers wouldn't rush to 64GB, they'd say " Cool I can multitask now at 256" or " great I can do horizontal scalability' or something else.
But no they created the market conditions so now what would happen is the consumer will immediately flip the 192GB they don't need on eBay, hoping to snatch a profit before the prices tank and the second hand market will be flooded the rug will be pulled out from the luxury pricing and everyone will get screwed.
This has happened in electronics markets before. Many times.
When Engels talked about the grave diggers of capitalism they were looking at it through a 19th century labor/manufacturing lens but arguably this same dynamic is at play here.
If you were a DRAM manufacturer, isn't this exactly the kind of thing that would make you think twice about investing years and $billions in new fab construction?
Let's say ram used to cost $100 and now that same unit costs $1000. You paid say $500x1,000 for that unit during the price increase or some price where you can currently flip for profit.
You have a very expensive data center and you're in debt financed on the premise that you have these special computers.
Now a new technique comes out and it turns out you only need 1 memory unit for something that used to require 8 or 4 or some meaningful multiplier.
This stuff happens all the time. It's why we don't use BMP files on websites or serve giant MOV files on YouTube. It's why postgres queries are faster now than they were 10 and 20 years ago.
You rent out your machines. You need to service your debt.. Demand may 8x overnight to accommodate but you have a monthly bill to pay and that's unlikely. It's likely going to drop.
Think about it. Your customers are paying maybe $10,000 a month and serving their customers. Now they can drop that to $1,250.
On market if you were to sell some of that ram you have 100% profit right now but not for long.
Jevons paradox assumes unlimited capitalization, zero debt servicing, infinite time horizons...
We live in the real world so what do you do?
Historically the answer has been "sell that shit"
There's an aphorism for this "stairs on the way up elevator on the way down"
If we had a healthy market with sane prices where you can't flip the thing you bought for 100% profit the answer would be "create more value."
Why not? Unlike many other workloads, LLM inference actually seems pretty suitable for decentralization (effectively stateless means no availability concerns; bandwidth and latency are relatively forgiving too).
People who say they want local runs really mean it: they want local runs on hardware in their room, not on some decentralized system which, if it existed, would almost certainly just be a worse, less-reliable version of cloud hosting. I'm not saying nobody would use it, but it sounds a lot like things like IPFS, which have also completely failed to displace either cloud storage or buying a bunch of disks for your own private use.
wonder what voting would be like?
gamer vote ++
datacenter hater vote --
datacenter lobby ++
micron lobby --
I know it's just a figure of speech, but damn. I laughed out aloud in public just reading this.
I guarantee Micron & friends are not intentionally orchestrating their business such that they would suffer a massively reduced chance of yielding on a per-die basis. Unless someone is actually buying HBM devices, they are not going to be making them. These are not a commodity that can be speculatively manufactured in any economically rational way.
Is it really so hard to believe that RAM prices are up because demand is simply exceeding supply, especially in a market where additional supply takes years and billions of dollars to come online? There's no need to posit cartel behavior and a fair amount of evidence that there is none.
I got a 4090 in 2023 for 1600, a 5090 in 2025 for 2000 with 256 DDR5 for about $1,000 ... and then, after some protectionist legislation passed, these prices quickly shot to the moon.
Connect the dots.
You can't say "connect the dots" at the end of a raving, mostly-incorrect post and act like you've made an ironclad argument.
Then good portion of those weights are n-grams (~200GB) that don't need to be in VRAM.
Then KV cache of that model is super lightweight at ~1GB per 1M tokens. If HBF succeeds, then accelerator with 16GB of VRAM and 1TB HBF/NAND is probably all you need (?).
Hardware update cycles are 2-3 years even on the high end, so it's still a ways away before "good enough" and "local" belong in the same sentence for the average person.
And by then, DeepSeek V6 Flash will be too cheap to meter, 5x faster, and 10x better, so... You'd still need to go out of your way.
Most people are spending most of their time on their phones anyway. ..
Edit: I went and checked for you. The LM backbone is 307.2 GB (286.1 GiB), straight from DeepSeek's upload. The n-gram table is 203.1 GB (189.1 GiB), which goes in host RAM. Note the embeddings are higher precision than the expert tensors, so it's a larger fraction of the bytes than it is of the parameters.
So,
> Call me crazy but:
You're crazy. :-)
1070ti launch MSRP was $450 ish. 5070 could be had in the last year for 5xx-6xx range easily.
All things considered - (inflation being about 30%~ (guess)) between these two timelines. You are looking at 300% performance difference at a cost dollar for dollar that is cheaper then when they purchased their cards.
Might be a bit of a stretch blaming it on "severely overpriced for too long..."
Allow a question from someone who’s only got a very vague idea of how this kind of stuff works behind the scenes: say I rent usage of this model through one of the many LLM hosting providers out there, and let‘s assume I use it extensively through something like Pi or OpenCode and vibe code away all the time, keeping the hosted model occupied as much as I can, happily burning my credits.
Does that mean that there is a hardware cluster as described by you above that is crunching away just for me?
So at FP16, I alone keep a 1,664 GiB system occupied all the time?
As background: For the most part VRAM oversubscription/paging/swapping isn't a thing in the same way that RAM for a VM often is. There are some approaches to it, but (to my knowledge) not at that sort of scale.
There are some systemic reasons for this, but very broadly speaking the GPU vendors are building toward the highest bandwidth and lowest latency possible, and the overhead/complexity of something like protected memory modes serves neither of those priorities.
It has a set of n-gram tables which you can stream from system RAM or even NVMe
That said it’s still quite big! I can’t fit it on my DGX Spark, though I believe you can if you have two?
I’m quite spoiled with how good Qwen 3.8 Flash Next is on a single spark though: shocking how good local models are getting on attainable-ish hardware
Because it's an open model so providers compete on price.
He gave demand signal so many times years ago and was mocked for it and now we have the consequences of industry not taking him seriously.
I reimplemented most of the features of the Deepseek v4.1 flash paper (apart from quantization aware training which doesn't make sense because my implementation uses float32 precision anyways)
I'm currently learning how to distill reasoning traces (check my other github repositories) but I think that a locally selfhostable deepseek is possible with my mixture of experts sharding mechanism. I decided to optimize everything for CPU parallelization, with the idea that the KV cache and meta model have to run from CPU RAM anyways, so the experts can also be loaded/unloaded at runtime if needbe, to save more RAM.
My assumption is that the KV cache optimizations in combination with the CED and compressed attention features are the reason why v4.1 flash has so few hallucination problems and such a strong self-lookup/thinking behavior. But that's more a gut feeling, need to evaluate and test this more thoroughly.
Anyways, would love to see someone train this on their own datasets. Currently my pipeline is kinda optimized for parquet and zim files.
No, you won't get frontier-level intelligence on a 1070Ti. Yes, it should be illegal to do what Altman did. Since we clearly don't live in the best of all possible worlds, we need to settle, and DS4.1 Flash is a good place to do that.
For tasks that don't require vision I personally like the NVFP4 quant of GLM 5.3 from Local Inference Lab better than DS4.1F, but they are both well beyond awesome.
1660 ti, 4790k, 16gb ddr3
I don't think those subscriptions nave negative contribution margins, either. I think we're seeing a lot of price discrimination by the big labs, and huge margins on their frontier models. The fact that they have been cutting prices to their second-biggest tier of models (Opus/Sol).
Open models catching up and collapsing these margins would worry me if I were a shareholder in the big labs, but as a user, I really doubt that the western labs have bigger environmental impact just because they have higher API costs, I think they have a ton of efficiencies they aren't sharing with customers yet because demand is so high.
Plus you can also get dsv4.1f subsidized. OpenCode Go gives 4x if I understand their pricing correctly. Anecdotally, I feel like I get way more out of my $10/mo OpenCode Go sub for the price than my $20/mo ChatGPT, even using gpt-6.1-sol high which is very cheap, and I have yet to convince myself dsv4.1f is a worse model.
It blows frontier API pricing out of the water, but again, look at cost per task, not token usage. Still easily wins though for my work.
I do think it's the most viable alternative I've seen so far, and that applies pressure to the frontier models. Should subscription prices hike or become unavailable for some reason, I know what I'll be using.
When pricing this, it's important to consider whether or not you want to opt out of data training. You won't get the advertised rate. Also the dsf 4.1 subscription providers are throttled af... and of course they are, because otherwise they'd be haemorrhaging money.
DS4.1 Flash not really cheaper than frontier models???
It is insanely cheaper.
DeepSeek is horrible at grilling sessions (the /grill* skills to make technical decisions). It doesn't know how to explain things. Maybe the skill could be adjusted. It also doesn't come up with as good solutions as Opus/Sol.
What I use it for is
* the orchestator of my coding workflows
* the tester/verifier of code changes
* the sub agent that explores code or does web searches
* putting together code base research reports
They are all working together in Pi using the extension @tintinweb/pi-subagents where my workflow skill is calling different subagents that use different models.
Luna is cost competitive, but doesn't score as well on intelligence. I do need the intelligence for most of what I use it for, so I am not motivated to use Luna. Haiku also doesn't seem like a competitive price/performance mix.
AA has Haiku 5.5 as cheaper than 4.1 Flash (both on Max, which isn't ideal but what can ya do) and a 4 point intelligence gap.
Why do people like to think open models are more competitive than they are?
I was previously using GLM-5.3 as the orchestrator, after switching to DS anecdotally there was an unnacceptable quality loss, mostly around not taking all the relevant context into account when making decisions, pulling new design out of thin air without discussion too often, and being way too wordy and rambly in documentation despite prompting to avoid it. There's a lot of docs, rulings, core concepts, design philosophy to uphold and DS was just not cutting it.
However, it's perfectly capable of being the sole agent for all of my well specced implementation tasks. I've gone back to GLM as the orchestrator.
The sub-agent separation is still valuable to keep context clean for the orchestrator, but I just have no reason to use Sonnet as the grunt-work implementer because I'm finding it hard to run out of tokens with Opus 5.5 on a $200 subscription plan. It's really really good at subjective quality of work per token used.
Once its gets juicier, we let flash launch specialized subagents with specific models. GLM-5.3 for coding or Kimi K.3 for research and critique.
But as a main driver. I love flash. And it brought our bill down by A LOT :D
(And what are the preferred providers?)
are you worried about sending all your data to third parties, especially if they're in different countries?
The model engine provider might be ZDR, but the service as a whole isn't.
nobody here is talking about running frontier level intelligence locally so if you’re Chinaphobic and prefer layers of corporations siphoning your data in between you and the party there are plenty of options instead of directly to the party
> Today's models are now good enough for high-quality unattended tasks. Chasing the latest and greatest is silly. It is fun to see the new Fable capabilities, but the tasks we throw at them are usually ridiculous (maybe even insulting) if you believe in LLM sentience. It's like asking a math PhD to organize the files on your desktop.
I'm using DS V4.1 Flash as my main model since their release and it works great for all my coding tasks. My setup is OpenCode Go subscription and obra/superpowers skill.
The only times I try to change models are on general planning tasks (like research this codebase for tech debt mitigation opportunities) or if I need deep research which would benefit from searching the web, in which I still think Gemini is still the best because of the speed and access to google search index. But these are not even 20% of my daily tasks.
but DS 4.1 Flash is good enough for most tasks