I was skeptical, but this looks genuinely useful. I can already think of some use-cases for this, e.g. a spreadsheet I was working on recently had a list of user, and then a comma-separated list of software apps they were using. It was a bit of a nightmare to parse, but the example in the article would definitely help with filtering the list.
I'm confused - is this meant as satire? The blog author's other posts also look like they are generated by an LLM, so I'm not even sure if the author is trying to be funny or serious?
Strange - I had the opposite experience. The more I learned about Jev the more interesting I thought it was. This Syntax video got me excited about developing with Jev: https://youtu.be/QbYBRjOaGOo?si=Up0QbW8tWCZQzZR4
No mention of the performance of the models? I'm able to load a bunch of different models on my little mini-PC with 16GB RAM, but the performance is terrible. I always wonder what performance people are getting with local models that they find is acceptable?
I run a similar setup to the one he described on similar hardware. I run bifrost and llama swap though (tailscale rocks). My local model usage is for some out of band batch processing one of my personal apps uses. Basically a personalized recommender for media, it curates stuff for me based on a database i've compiled over years, so non-interactive. For that use case, I don't really care that it might take a few minutes to run. It's free. The machine is just sitting there anyway. I have tried using qwen-coder and opencode on my M5 Max 128gb and compared to claude code it's painful. I did setup a workflow where claude plans, qwen executes (unattended overnight, again b/c it's slow) and then claude reviews. I benchmarked this several times and I ended up using MORE tokens with claude because it had to 'fix' all the qwen issues. While the code it produced was 'good enough' the fixes were worth it so I just stick to coding task using API models (codex and claude).
It isn't, the cost is included in your electricity bill, not even talking about the cost of your time to set it up. It's very possible that it costs you more than a cloud mode would, you just don't want to calculate it properly.
It sounds like they are doing something similar to what I described in my other post below. Personal media station.
That can be done on hardware that quite a lot of people basically just have and don't use 24/7 to the max - because it is their gaming machine or their programming and compiling workhorse, for example. Of course you are paying for additional electricity but even with napkin-math instead of a "proper" calculation, you are unlikely to pay more for running your own instead of something commercial (and that can be offset further with some of the "modern" electricity contracts and/or PV and battery storage). Especially if we are talking about a stack that runs most of/all the time when you are not using your machine and makes LLM calls regularly while running.
The work in software/admin to get whatever you want set up is similiar no matter which infrastructure you use.
true but if you're actually running k8s and similar workloads, chances are it might eat memory that LLM requires.
you'll also notice these articles rarely specify their context window in tokens, because it is small, usually 30k to 70k tokens and it gets slower as it fills up.
I actually have a Mac Mini M4 Pro with 48G. I gave the k8s example because this is what I was doing with it.
Was because I am back to using Linux as my workstation.
My Mac Mini is now a headless server for llama.cpp.
So, you are right that for these workloads , I would not be using the Mac Mini for k8s AND llama.
Another thing going against using a Mac for Linux containers is that there are no solutions that I know that properly manages memory : memory is given to the Linux vm , but never fluctuates if the needs in the vm are less than the initial request.
I know Orb Stack does that but is it proprietary. I think UTM does it , but not sure I would use UTM instead of Lima, Colima , multipass , etc to run containers.
> It's very possible that it costs you more than a cloud mode would
...which is almost always true in a single request/reply mode and never true in batch mode. Single request usually 2x-3x more expensive than cloud and batch mode 2x-3x cheaper. Now, for narrow tasks, a finetuned tiny 8b model would dramatically outperform SOTA frontiers for a fraction of price, esp. on energy efficient hardware like Apple.
Local is never cheaper than cloud because they can do batch inference, and that means you load model weights once to produce 128 tokens on 128 sessions in parallel not 1 token on 1 session like local models. Local models rarely get to high utilization factor, they spend most of their time waiting.
If you had only batch inference and enough of it to fill the compute to 80% then you get cheaper local models.
Local models can absolutely run in batch, what are even talking about?
> If you had only batch inference and enough of it to fill the compute to 80% then you get cheaper local models.
Even if you ran sequentally, single session, a _finetuned_ tiny (8B) local model on narrow tasks would abolutely mog SOTAs, any of it - Fable, Opus, Sol you name it.
> Local models can absolutely run in batch, what are even talking about?
I think the point was that if you aren't running your local machine at 100% for 24 hours a day then a cloud - with multiple clients - that is, will be more efficient.
Can you share a bit more about your bifrost and llama swap setup? I’m facing memory constraints and am looking for a managed model solution that will help with hot swapping loaded models and stay-warm concurrency. Ideally with prioritization.
What do you want to know? Just start llama-swap with the models i have downloaded, add llama-swap as a provider in bifrost, expose the models you want and they become available in one single endpoint you can use in anything like opencode, openwebui or anything that speaks openai.
Yes, that makes sense. Some of my models currently run in ollama while others require their own inference servers. I’m curious about custom inference servers in bifrost and the ability to orchestrate keeping some models warm in memory while evicting and cycling other models. All of which span different providers in bifrost (I think). Obviously I can get the mechanics from an agent, more wondering about any experience with something similar.
i have a 512gb ram m3 ultra mac studio setup with a gas city that runs one of my companies. today was the first time ever that a local model (GLM5.3 8-bit) was able to match fable5 in our tests.
GLM-5.3-Flash at true 8-bit: 341 GB on disk, 328 GB resident, 288 experts across 46 layers, loads in 65 seconds.
• 18.7 tokens/s generation, 35 tokens/s prompt, on a desk, on a $0 per-token bill.
• Runs beside our whole agent city on one box with ~130 GB to spare.
• Review test: caught 6 of 6 planted P1 defects, zero false positives, same score as the frontier model we pay for.
• CRM test: 11 of 11 required records extracted, zero wrong writes, 45 minutes, first local model to clear the bar.
• Serving a 131k-token window today; the model itself supports 1,048,576. Widened to 4 concurrent slots and still have 50gb+ of excess ram.
granted my cto still isn't moving all of our inference to glm5.3 but we've identified 40%+ that is currently handled by fable that we're routing locally instead and will do concurrent requests to verify/compare responses for a while.
You still have electricity and capital investment. Envelope math suggests cheap electricity is costing you something like $0.50/mtok and the opportunity cost on the capital tied up and lost in the unit purchase and resale is going to cost you something like $2/mtok at 100% utilization (so, frontier model prices or higher at real utilization), and you don't benefit from any elasticity.
Hosted GLM 5.3 flash is like $0.15/mtok in $0.50/mtok out
Time to completion also must be considered. If I have to wait around for hours for a prompt to complete locally and I’ll need to iterate quickly, I’m better off hosted than local. If it’s “free” and slow it may just not be worth it.
Props to your parent commenter for including context size. Because 131k context window is prohibitively small for my coding workloads so I know a 512GB Mac won't cut it.
I have an M4 pro (48 GB ram) and I run Gemma 4 26b a4b at 52 tok/s and Qwen 3.5b a3b at 72 tok/s. Both 4bit quantized. These are enough for my needs and the performance is more than good enough. I'm not running the MLX version of the Gemma model, if I did the inference speed would likely be a bit better. I wouldn't use them for coding features though.
My perf sucks compared to yours. Added it to the post - same model averages 325 tok/s in processing prompts, and 34 tok/s in token generation. What am I doing wrong..?
Wow, that's just about half the perf. I'm not sure what you're doing differently, though our hardware is a bit different: I am on a Macbook Pro M4 Pro, while you're on a Mac Mini.
I would try a different version of the model from HuggingFace while ensuring it's MLX. I'm also using LM Studio, not oMLX, and I've seen some threads like these:
Some examples (keep in mind this is all indefinitely free for me, no burning quota away):
1. Getting information (such as information about hardware unfamiliar to me) when not connected to the internet, which happens occasionally in my case.
2. Continuing to learn Rust by way of toy examples, puzzles, and comparing aspects of various solutions, for example from LeetCode.
3. Reformatting data, for example from a PDF to a markdown table, or converting receipt images to text.
4. Simple translation/explanation (e.g. I'm teaching my wife one of the languages I speak but sometimes may not know/have the words to explain the full nuance of a translated word).
5. Summarization. One of the webnovels I'm reading has some very boring parts I don't want to slog through, in those cases I simply make the LLM summarize that part and move on.
Etc., you get the idea. It's not unusable for coding, but it would make many mistakes when making a whole feature and the context lengths are limited to around 30k-40k tokens by my RAM. I could give it access to the web but I simply use an online model when I need that sort of thing, again partly due to the context limit.
Edit: The MLX version of Gemma 4 26b a4b does about 62 tok/s.
I'm the author - hello! Added to the post! Qwen averages 325 tok/s in processing prompts, and 34 tok/s in token generation. That isn't instant, but it's quick enough that I never really think about it.
It’s not. Do it as a hobby or for privacy but for performance just use a frontier model api. You’re paying less than cost for something that would take tens of thousands to set up locally.
That's not even remotely close to being true, even once you account for capex. You have to look at the actual usage, look at the token limits. Even if you're paying Anthropic $200k/month for scale-tier, you're going to blow through your token limits trying to run max output 24/7. Three users running Opus 4.8 at max non-stop will probably clean your monthly allowance from daddy Dario in less than a week.
With an 8x MI355x cluster at full tilt and including cooling, your power draw runs ~17kW. That's what it looks like when it's running full tilt. To be fair, hey that's pretty expensive. It does mean 8 multi-trillion parameter models unquantized running 24/7 without pause. And you get the full month like that, your monthly token limit is the time in a month. That cluster, the electrical upgrade, the cooling setup, and the electricity to run it all costs less in 2 months than your maximum affordance from Anthropic does in the same time period. Two billing cycles, and realistically it's more like two weeks. In 4 quarters you've wasted over a million. Like, what are we talking about here?
Now if you aren't using AI all that much, which is perfectly valid, and especially if you aren't using it at its absolute maximum, the story changes. Because even though at that point you're not paying nearly as much in electricity to run the cluster anymore, you still have the $300k+ capex to get the setup in the first place. But if we're not redlining it non-stop, then we're not really talking about performance anymore, are we? If your org never comes close to hitting token limits, it's probably because AI is rather marginal for you. Which again, is perfectly valid. I don't even use AI professionally.
Fact of the matter is, if your corp can justify the capex for a cluster and makes heavy use of AI, you are literally burning money by not having one in your building. The numbers are painfully obvious. Even deepseek isn't as cheap. This is before we get into things like LoRAs, custom inference pipelines, etc. which you know are kind of important if you actually care about model performance.
> With an 8x MI355x cluster at full tilt and including cooling, your power draw runs ~17kW. That's what it looks like when it's running full tilt. To be fair, hey that's pretty expensive.
Pretty expensive is an understatement. You couldn’t buy one of these if you wanted to right now. If you could it would be multiple hundreds of thousands of dollars.
> It does mean 8 multi-trillion parameter models unquantized running 24/7 without pause
You can’t even run one unquantized multi-trillion parameter (>=2T) model on 8 x MI355x with enough context for concurrent users. I don’t know how you think it’s going to run 8 of them at the same time. Did you mean 8 concurrent sessions?
Your math is way off across this post. If replacing an Anthropic subscription for a whole company was as easy as buying a box for the office and then breaking even in 2 months, it wouldn’t be some little secret that we only discover in a comment online.
> Pretty expensive is an understatement. [...] If you could it would be multiple hundreds of thousands of dollars.
Obviously, I quantified both the operating expense and the capital expense in my post. What I find curious is that you're quoting me talking about the operating expenditure, and changing the topic to be about the buy-in like these are interchangeable things. You don't think that this is a crucial and important distinction?
> You couldn’t buy one of these if you wanted to right now.
You could have spent all of 5 seconds of searching rather than just assuming[1]. You're not buying an Nvidia Superpod™.
> You can’t even run one unquantized multi-trillion parameter (>=2T) model on 8 x MI355x with enough context for concurrent users.
That's certainly fair a point. Although in the English language, especially in legal contexts, the multi- prefix is used inclusively for fractional values. That is it's strictly >1, not >=2. IE an 18 month contract is a multi-year contract, or a $1.6 million dollar asset is a "multi-million" dollar asset. But this is uninteresting semantics.
You are right, but it also doesn't matter. The gap is just that big. You can run 1 single user of Kimi K3 and still not even come remotely close to the $70k or so that a single Opus 4.8 user can burn over the course of a month on left on max. An honestly lowballed amount I know from anecdote. The per-token cost is just really expensive.
> Your math is way off across this post.
You made one technical point above, one that doesn't ever arrive at a relevant rebuttal to the substance of my post. But please, I'd love to hear you elaborate, especially because I didn't actually give much math at all.
If you want math though, here's the math. Let's say you are paying a ridiculous amount of money for electricity, a price nobody in the US pays -- $2 per kilowatt hour. That's about 5x the average rate in California, 4x as in Hawai'i. 17kW @ $2/kWh * ~8766 hours in a year puts that cluster's electrical costs at just shy of ~$298k annually assuming it takes no breaks. Let's make matters worse and round that up to $300k. It's also assuming you didn't invest in a solar hookup for your building, which I don't know why you haven't at this point, especially if you're installing a CDU for your new cluster. 12 months of Claude burning $70k a month is $840k. For a buy in of, you know what, let's call it $500k. Why not? It still doesn't matter. The operating cost is so much lower it's paid for itself plus an additional $40k in the first year. Even at a ridiculous penalty in electricity that nobody pays, even overinflating the amount of money you'd pay for the cluster and the infrastructure to get it set up, it's not even remotely close for a single user where the gap is smaller (IE, you're not wasting "a million dollars" in a year by maxing out the $200k scaling limit every month)
You can of course trot out the point that oh, in 12 months this setup will be extremely outdated! It doesn't matter. If the work it was doing today was useful, it will be useful next year too. And with the rapidly encroaching diminishing returns from parameter scaling, you're probably going to be just fine for a while. Maybe grab a quantized version of a newer Chinese model at the end, before grabbing a newer generation of AMD node. Those MI400s are looking pretty sweet after all.
> If replacing an Anthropic subscription for a whole company was as easy as buying a box for the office and then breaking even in 2 months
If you're locked in, then you're locked in. But don't pretend like you're saving money. You're not.
> it wouldn’t be some little secret that we only discover in a comment online.
Why does this have you so nasty and defensive? It's not a "little secret" that running your own infrastructure is cheaper. Of course it is. You know what else is cheaper? Owning your own office building out in the sticks, rather than leasing part of one in the city. Not everybody can make that work, there are no free lunches after all.
History repeats, these same exact lines were rolled out ad nauseum during the cloud craze. Datacenters are businesses, not charities. Frontier companies rent quite a fair amount of their infrastructure. Even if they resold that compute below cost (they don't), there's a pretty steep cliff before the economics start to look attractive.
> You could have spent all of 5 seconds of searching rather than just assuming[1].
I guarantee this will not ship to you any time soon.
The current lead time on these GPUs in measured in years. If you didn't place an order for this a long time ago, it's not coming this year.
Being able to add it to an online configurator does not mean anything right now.
> 12 months of Claude burning $70k a month is $840k
Your math is completely useless with these arbitrary numbers pulled out of the air.
If you want to begin calculating payback period you'd need to look at token costs, cost per task, utilization rates, and so on.
> The operating cost is so much lower it's paid for itself plus an additional $40k in the first year.
You went from paying back in a couple months to paying back in a year but you still haven't even talked about tokens or concurrency.
You're also neglecting the fact that hosted tokens are going down in price at a rapid rate. If someone was paying $70K per month in tokens for Opus this month, that same level of compute is going to be much cheaper 12 months from now.
> Why does this have you so nasty and defensive?
Not nasty or defensive, just tired of these armchair claims that it's easy to go out and buy an 8 X MI355X box from people who obviously have no idea what the hardware lead time is like right now, or who haven't considered the actual math on token costs and payback times. You're still making a lot of claims without a single discussion of cost per task or token.
It usually boils down to people trying to convince themselves that keeping their macs hot and with very little ram to spare only to get sub 50 tokens per second on a subpar lobotomized (quantized) model is worth it.
And I'm not even considering their time spent fiddling, fine tuning configs to adjust for ram, updating/benchmarking models, etc. Which is probably more expensive than the mac so the math is even more wrong.
> I guarantee this will not ship to you any time soon.
The assumption, the starting point, is that you have a line on the hardware. Asking around, some distributors have a 6 month lead time on Instinct GPUs, which curiously enough is about how long you'll be twiddling your thumbs waiting for the cooling loop to be put in. Yes things take time.
> Your math is completely useless with these arbitrary numbers pulled out of the air.
Your dismissal is worthless if you can't even be bothered to provide a counter-example. You've not provided a single iota of quantified reasoning beyond my original not accounting for the space used for the context of concurrent users.
> If you want to begin calculating payback period you'd need to look at token costs, cost per task, utilization rates, and so on.
Now go back and carefully reread my original post. Yes, if you are not actually redlining an LLM for a billing cycle, the capex starts to be way more relevant for this setup. Otherwise, our constraint is time and our unit of measure is $/hr.
If you want to compare token cost, it may shock you to learn that Kimi K3 without speculative decode on this setup is slightly under twice as fast as Opus 4.8 max. That's still true when fast is compared with K3 with speculative decode, and now Claude is twice as expensive as a base rate. Oops. We're already burning more money over a period of time, looking at tokens we're screaming even further ahead.
> You're also neglecting the fact that hosted tokens are going down in price at a rapid rate.
Cool. Call me when Opus 4.8 max is $0.50/million. In 4 years you could have bought the 200 acres of land down the road from your building, started a 5MW solar farm subsidiary that you'll expand over time, and as soon as your connect is up, dropped the opex of the cluster down to its maintenance costs. That subsidiary will pay the loan required to spin it up back irrespective of your primary business. When you own your own shit, you can play your own game, stack your cards deep. Have a little bit of business acumen. Fuck what The Valley is doing, that is an ecosystem fully enslaved by economic nihilism, money isn't grounded there.
> Not nasty or defensive, just tired of these armchair claims that it's easy to go out and buy an 8 X MI355X box from people who obviously have no idea what the hardware lead time is like right now
This motte-bailey routine is both nasty and defensive, particularly when you keep prosecuting a geist of numeric justification that never arrives. All I've gotten from you is vague dismissals, one borderline irrelevant technical argument, moving goalposts and missing the point. Granted, not as egregiously as other people in this chain thinking we're talking about running 100B models on a Mac, I'll give you credit for that. But this whole time, we're just talking past each other. You make realistic points and I try to bring you back to context, but you have to work with me here too.
The point was that these companies are not selling to you below cost, they're not even selling to you at-cost. Just use your head. Venture capital isn't a magic wand. Frontier companies are in the red because they're in non-stop expansion operations at massive scales. Anthropic has an operating profit of half a billion dollars[1]. They are not selling you API usage below cost.
"You can of course trot out the point that oh, in 12 months this setup will be extremely outdated! "
But interestingly still extremely valuable on the second hand market.
The capital expense isn't the amount laid out. It's the rental cost of obtaining that capital, less the depreciation on the fixed asset over the period in use.
Going back to the OP, Apple gear is well know for having good resale values, which means the capital outlay isn't anywhere near as much as some people think.
>You couldn’t buy one of these if you wanted to right now.
You can: https://www.exxactcorp.com/Exxact-TS4-149591758-E149591758 . You can get thousands of tps of GLM 5.3 output out of this thing, which grades around Opus 4.8. Payoff is around 1 year vs. spot prices on these GPUs, including power.
Ah thanks for the solid info, too bad. I'd seen them come up as a pretty good price for 6000 RTX's in the past, which seem generally pretty available, good source for those?
Yeah, they're good source. But the price for those GPUs is 5 figs even with the nvidia startup program nowadays. Also, I went back and looked. Most of my GPUs are actually from Central Computers who were great, but Exxact is real too. So "lots of" was inaccurate.
Also, the lead time I quoted was for individual 8x nodes.
I can't tell from the ad -- it says "supports" 8x MI350X GPUs, but does that mean "includes" 8x MI350X GPUs? For $300K I'd certainly hope so, but I'm assuming not.
A system with 4x RTX 6000s costs about $60K these days, and can (as you note) trade blows with Opus 4.8 if not Fable. In fact, it'll give you a better pelican than Fable 5.1, and in less time.
Well, they are if you're into animating pelicans. :-P But yes, in the general case Opus is a better match.
And Opus is no slouch. I'm satisfied that GLM 5.3 is just as strong as Opus. Z.AI has promised/bragged that they will be at Fable 5.0 level by the end of the year or early next year, and I don't see any reason to doubt them.
Baseline yeah. But part of the reason you run open models is how much nicer fine tuning them is. Granted, you probably don't want to try and make LoRAs on a 4x RTX6000 setup, but you could if you really wanted to and there are other ways to modify models. And yes, if you're good at it, you can turn a piddly mid-range model that's only good at benchmarks into a heavyweight clanker (for a specific domain).
Ha fair, I'd definitely confirm with a salesperson before wiring them $300k. But most of the signs on the configurator seem to point to it including the GPUs? Not going to make 30k BTUs/hr of heat without the 8kw of GPUs.
Here’s an experiment: purchase an anthropic pro max subscription for $200/m. Now go buy the hardware to run DeepSeek’s equivalent. In a year, who spent more?
In normal times in which hardware used to depreciate (lately that's not the case and HW even appreciates, but let's not get distracted), if you calculate only with depreciation costs, plus the fact that when you have such a setup, it'd take many 200$ subs to cover your lack of limits in the other, I think it'd not be a clear victory for any side.
If you just ask "who spent more in the first year" (100% depreciation) then even with 5-6 max accounts, buying HW will be a couple of times more expensive. But when does it make sense to ask that question?
Maybe the SotA models will need better hardware so your investment will not be useful after a year or you'd need very expensive upgrades? But then (as in Fable case) subscribers need to spend more too.
It’s not so clear after 5 years that you’ll come out ahead. You’ll have spent $20k. The apple computer owner will probably be running local models that are better than today’s frontier on the same hardware.
Idk where you live, but where I am running the M5 Ultra Mac Studio at max rated power 24/7 for a month costs C$42.
The considerations against Apple hardware are 1) hardware advancements 2) early access to the best models. But it’s really not that clear.
(The other guy who thought hosted models on openrouter are cheap has spent $100k in 5 years.)
Open weight models have been getting better/smaller every year.
Also, from what I can tell, MLX inference is not as well optimized as CUDA, and the M5 Ultra has additional kinds of AI compute which is unavailable on other M models. With the massive 1.2 TB/s 512GB Mac studios coming out, I think MLX will get a lot more attention.
In short: Todays models should run faster next year, and next year's models should also be more efficient.
It does make me wonder how the hosted stuff is so cheap. For pretty much everything else, hosted/rented is more expensive but offers better convenience and flexibility. But for AI, even if you consider the total lifetime cost and are utilizing it heavily. You never break even by buying.
They're not cheap at all. I did one xhigh Qwen 3.8 27B agentic coding task last week via OpenRouter and it cost me like $10.
99% of the cost was in input tokens, I only used like 100k ish output tokens. It was a one shot task asking the agent to implement proxy injection to Guice. It did a pretty amazing job.
If you were to use hosted LLMs for a lot of agentic coding, a maxed out M5 Ultra Mac Studio would pay for itself in under a year.
I've been hosting Qwen3.8-27B myself. On my endpoint it's $0.30/1M in, $0.10 cache, $2.03 out - so those agent turns that re-send the same prefix get a lot cheaper when cache hits. UI at inference.tiyuvta.ai/app if you want to try it. Hosted is up to 210 tok/s and 280ms TTFT with reasoning off.
Qwen is weirdly expensive. Deepseek v4 flash is dirt cheap. You'd need at least 128gb of ram to run this model and in my experience, a days work with it costs around 80 cents.
So I ran the math, assuming the agent takes 75 turns per 200k context, with deepseek v4 flash it costs around $2.57 to reach 1M context in 375 turns. Cached input costs scale quadratically with # of agent turns.
Considering that I hit the 1M compaction multiple times per day with codex, it would definitely cost at least $5-8/day to use deepseek how I normally use codex.
Here's the thing that's a little different about data centers; we can tell from Anthropic and OpenAI that they're capacity constrained. Inference demand is there. I notice Cerebras doesn't offer much directly any more, all their capacity is getting completely sucked up by B2B sales. Grok did overbuild, but Anthropic was so desperate for more compute they ate their pride and leased the excess capacity.
That means all these data centers are being heavily utilized by actual end user inference demand. Well, some is research on new models, but a lot is actual end user demand. No one has given an explanation of why peoples usage would decline.
On top of that, margin on inference appears to be decent. It's model training that's a serious financial burden.
And maybe that's where there will be a slowdown, maybe the market doesn't justify spending as much on R&D as it does, but the end demand for inference is there.
Does that justify these stock prices? That's a different question. But the housing boom left behind endless rows of empty homes because demand disappeared. The 'dot com' boom left behind thousands of miles of dark fiber that'd been built out well ahead of demand for bandwidth. I can see the stock market having a giant sell off, but I don't see data centers sitting idle in that same fashion.
The number of planned data centers and the scale is pretty nuts.
Here's what you have to believe:
- AI demand is at least several times larger than what can currently be satisfied, or will grow. (This one I can buy, but...)
- AI chips (GPUs, TPUs, compute-in-memory, whatever else is being studied) will not get significantly more efficient than they are now. It will not be possible in, say, 5-10 years, to do 2X or 4X or 10X more AI requests per rack than is possible now. I think this one's the single most likely thing to be false, since all computing history contradicts it.
- Edge devices (PCs, laptops, specialized but smaller scale AI compute nodes) will never be powerful enough to run frontier models at a reasonable price that's appealing for professionals, enthusiasts, or businesses, and there will never be a market for this. None of the demand will be served on-device or near-edge. AI must all go in giant data centers.
- AI models will not become significantly more efficient than they are now. There are no large gains on the table from better model architectures, better training, more efficient quantizations, better harnesses, etc.
If all those things are true, than the current planned like 4X-10X increase in data center capacity makes sense. If even one or two of them are not true, then the planned data center build-outs start looking excessive. If all four are not true, it's a total bubble that will crash and burn. Answer is probably somewhere between, but how far toward bubble? That's why I picked a number like "only 20% ever gets built." It might be as high as 50%. It ain't gonna be 100%. The planned built-out is batty.
Oh I forgot two more...
- Data center capacity currently serving non-AI work loads does not shrink through either reduced demand, more efficient software, or (most likely) faster chips and denser RAM. If that happens, more pre-existing DC space can serve AI work loads.
- Orbital solar powered compute nodes never happen. If this happens (free power! much less political opposition!) then terrestrial data centers have significant competition.
I honestly wouldn’t bother with local models right now unless I either had a 5090 and was happy with running Qwen 3.8 27B, or a pair of DGX Sparks running DSv4 flash, or better, 2x6000 RTX Blackwells. Those are the kinds of rigs that the local model enthusiasts are running. With the GPU setups, you’re looking at generally >100tps generation in single stream, and >10k tps of prefill, so it’s snappier than Claude code, which somewhat makes up for it being dumber.
That said, it is really cool to be able to run an LLM on eg a Mac laptop. Just not a better experience on almost any metric for interactive use than eg Claude Code, beside privacy and guardrails.
>I honestly wouldn’t bother with local models right now unless I either had a 5090 and was happy with running Qwen 3.8 27B
How's the actual performance of Qwen 3.8 27B? On deepswe it supposedly performs slightly worse than gpt 5.6 luna high[1], but I can't help but think they've been benchmaxxed.
Not sure, I haven't run it, I've just been running DS V4 Flash non-stop since it came out, and that's replaced a lot of my Claude Code usage. People seem very impressed, though, it seems like it trades vram/world knowledge for extra thinking time, which I think is a good trade for local. tbf, I've heard luna's not great at coding. Fast and good for things like classifiers, summarization, though.
A friend and I were actually discussing today how benches show Luna Max at about par on coding with Sol Medium, but how it's nowhere near in reality. We were speculating that maybe it's because a lot of benches are best-of-n, and should probably be worst-of-n, because variance in performance is killer with large coding projects. Consistency is what lets you actually build on this stuff.
I took this thread and summarized it with Qwen3.6-35B-A3B, it had 1400 tps prefix and 60 tps completion. Very good performance. Using oMLX on MacBook M5 Pro 64GB.
Are you using the right configuration for your own CPU?
On a Laptop with 32 GB RAM and Iris Xe integrated graphic card, I get between 11-18 Tokens/Second with Qwen 3.8 27B and llama.cpp with sysl Intel optimisations. Same results with the vulkan back end, although sometimes it ends in weird segmentation faults due to the memory consumption.
I agree with Mitchell's original post: "stay frosty with AI", and I disagree with response that says "the houses are better". I'm assuming that Mitchell was saying you should treat all AI code with caution, and the more I work with AI-generated code, the more I agree with this sentiment. Even the frontier models (Fable, Opus, Sol, etc) all make fundamental errors from time to time, and if you're not keeping your eyes open these are easy to miss.
Just the other day, I got Fable to help generate a sign-up flow for a Django app. It looked perfect to me, and like the Twitter post linked to, I looked at it with marvel and delight. But I got Sol to review the code and it spotted some fundamental issues with the flow that would have let bad actors enumerate accounts and log in before their account was verified - both of which were supposed to be checked.
To continue with the analogy, yes we can "will houses into existence", but unless we've checked the bricks and the foundations, these houses can easily crumble into rubble.
Here in NZ, what Trump has been doing in the Middle East is having a huge impact on policy, especially with the upcoming elections this year. Trump has reminded us (and I imagine many other countries too) that relying on other countries for fuel is fraught with danger. EV sales have spiked here, and all of the political parties are scrambling to come with electrification policies. So for NZ at least, I definitely agree with the OP comment about Trump's influence.
I was initially surprised that Gruber was so invested in the "quality" of AI-generated text, which in my mind is an oxymoron. But really, Gruber's interest here is with the EU. This forms part of his ongoing attacks on the EU, all because they have been forcing Apple to align with regulations.
Can we install random unapproved apps on our iPhones yet, or is Apple aiming to just be fined a trillion dollars because they make more than that from the 30% cut?
I am from EU. Alas it has a tendency to produce some idiotic regulations. Cookie banner, new packaging fee, etc. I genuinely think some Apple related ones hurt customers more than help them.
Just a regular reminder that cookie banners are not mandatory. In most cases, they are a case of malicious compliance. The choice is between (a) not doing shady stuff with people's information, or (b) having to tell people about the shady stuff you do with their information. That almost every website chose to go with (b) seems to be a problem that goes beyond the EU regulation.
Even the website looks AI generated. Lots of cards with hover effects, icons in square boxes, the button shapes and behaviours, and just the general look and feel has Claude written all over it.
Nice - I've been considering re-subscribing to OpenAI to try out their new models. This is just the push I needed. Looking forward to trying it on Fedora 44.
This is very similar concept to "0h h1" https://0hh1.com/ which I've been playing every day on Android for a couple of years now. I had never considered it to be a reverse minesweeper.
reply