This will be interesting for a few reasons. First, depending on where the median pricing settles w/ 3rd party providers will tell us what it costs to serve a 3T model. Since it's going to be mxfp4 native, it'll take ~1.5TB of VRAM to host this, which is juuust at the limit of 8xb200s (but realistically you'll need 16x for context / throughput optimisation). Won't be cheap to host, but at least we should get some range of $/MTok for a 3T model. Then we'll be able to guesstimate if "labs are subsidising tokens on API pricing".
Also interesting to see what effort it will take to fine-tune this beast. The latest AISI benchmarks on cybersec place it above glm5.2, but still way way behind SotA closed models. Some fine-tuning might be needed here. Also, interesting to see if Cursor does another training round on it, to directly compare it w/ kimi2.6/2.7 fine-tunes (composer series) and grok4.5.
Also also, interesting to see if someone takes on distilling (proper distillation, w/ training the entire distribution) from this into smaller models. (dsv4-kimi should be really good, since dsv4 is very cheap to serve)
It will be very interesting to see what kind of 'slow' performance people get from running it on a no GPU, but tons of RAM server (like a dual or quad socket xeon with 1.5 to 3TB of RAM). For the purpose of giving it longer duration tasks to generate a piece of something and come back and check on what it has done in 4 or 6 hours. Even if the output is like 5-6 tok/s, that might be usable for some purposes.
Huge price difference in what you can do with buying a used 4U rackmount server and putting 3TB of RAM in it (64GB DIMMs x quantity 32 in a quad socket xeon, you can see some benchmark prices on eBay for sets of 16 or 32 matched 64GB ECC DIMMs) for <$30,000, vs the cost of trying to run it on real GPU hardware.
Now obviously, as of the time I write this, the full precision hasn't been released nor has anyone like unsloth run it through quantization yet to produce a "Q8" or "Q8-XL" variant of it. But I think it's going to need more than 1536GB of RAM, with a usable and large amount of context, more like 2TB and preferably 2.5 to 3TB.
I also predict that people who try to run it in Q4 and Q6 will get the worst of both worlds, less precision/lost knowledge but also not reliable output that comes out too slow. In my personal opinion if I'm going to deal with something that is smart but slow and running on limited budget hardware, I need it to be Q8.
> Even if the output is like 5-6 tok/s, that might be usable for some purposes.
You'll spend ~100x more on electricity than the API cost to have it run on someone else's GPU at several hundred tokens per second.
I think some sort of extreme data privacy requirement is the only situation that justifies this, but the intersection of {needs absolute data privacy, needs to run SOTA model, cannot afford GPUs} is really really narrow. I wouldn't be surprised if this is an empty set.
There are a number of use cases where sending the contents of your context and prompts (and the resulting output) to a 3rd party service is off the table as an option, and people will compromise speed for data sovereignty. And not everyone's electricity is equally expensive, I pay about $0.075 USD per kWh. It would for example cost me about $48 a month of electricity (not counting cost of cooling) to run a quad socket Dell R940 for a month.
That's an unusually low electric rate for the US - way below the lowest state average which is Idaho at 12.4 cents. It's certainly possible that you are getting 7.5 cents including delivery, but I've had friends say that they're "getting 13 cents per kWh" here in Massachusetts, but that's just the supply rate and the delivery is another ~18 cents.
There are parts of states like Grant County Washington that have cheap hydro power, but it's very rare for power to be that cheap in the US. Even if this applies to you, it won't apply to the vast majority of people on here who will have electric rates 2-4x higher.
Average electric rates by region:
New England 28.1 cents
Mid Atlantic 25.1 cents
East North Central 20.8 cents
West North Central 14.8 cents
South Atlantic 16.1 cents
East South Central 15.5 cents
Mountain 14.6 cents
Pacific Contiguous 26.1 cents
Pacific Noncontiguous 42.1 cents
>Do you actually need to run the state of art model at 5 tokens per second instead of a qwen or whatever 7b or 30b model at 100 tokens per second?
Some people like doing things they want to do. Do I actually need to buy expensive pigments from europe to make paintings of flowers? My camera produces a much more accurate representation.
Very good description of it. It does seem like a bit of a rhetorical question to ask a forum that has a very high population of Linux and BSD users why they might desire to have the option to do something themselves rather than relying on an external packaged ready to go product.
Do I really need to? No, not really. The 27B full density, 35B MoE, 70B and 122B models I have in use get me 95% of the way there on a lot of things. Particularly when dealing with languages and systems where I have at least an intermediate level of knowledge on, to know whether something is going down a dead end, using a wrong method, metaphorically chasing its tail, or is producing valid output.
On the other hand, would it be cool to also have a really big thing as an ancillary tool that I could throw a request into opencode before going to bed, let it crank away and take a look at what it's done 7 hours later? Yeah, particularly if I (very much an unknown quantity at this time) could be confident that it builds high quality, syntax valid, appropriately commented and not absurd code.
As someone who has worked in two industries that are at the maximal end of data sensitivity and privacy this comes across as a tinfoil hat issue not a real business requirement. In such cases we've always found ways to trade dollars for the privacy we need without having to run our own inference at excruciating slow speeds.
Do you mean by trading dollars for the privacy you need as:
a) Contracting with a third-party independent inference provider who will run your choice of model on fast hardware that they own, with all appropriate data security/privacy/contractual/compliance protection in place
or
b) Contracting with the original creators of the model to run inference via their API and with assurances that all the same data protection is in place
or
c) Spending the money to buy your own inference hardware to run it on something you fully own/control at proper usable speeds?
Edit: Everything I've been writing in this thread is mostly within the context of being able to evaluate K3 and its usefulness to be self-hosted as a preliminary proof of concept or test of feasibility of a new thing, such as on <$20,000 of server hardware, before proceeding to spend 300-400k on GPU-related hardware, or external third party services/ongoing billing.
They'll give you HIPAA compliance, they even have a data center for US government classified data, they can give you European data sovereignty. And with OpenAI and Anthropic models to boot, you don't even have to settle for open weights.
What kind of privacy needs do you really have beyond that?
It is not my use case but given recent political developments in international relations caused by the executive branch of the US government, off the top of my head, I could think of a lot of European or Canadian firms for which that would not be an option. No matter what they might promise about European sovereignty. For a good 'ol patriotic US domestic company? Sure.
Yes, its the US cloud act risk EU companies run up against on hyperscalers like MS/AWS.
Even for EU companies running open weights on EU stacks LLM inference on the GPU must process plaintext and I can't find any EU provider with NVIDIA H100/H200/Blackwell CC mode plus SEV-SNP or TDX, where you can cryptographically verify the workload ran somewhere the operator cannot inspect.
Personal compute is therefore the only option if you want personal autonomy privacy for IP &c. Maybe another option is to use cloud compute rented to fine tune a personal model that suits your own needs that would help bring the cost down, I don't know enough about this area to know if it kills the "intelligence" of those domains due to limited ?cross-verification within the LLM.
There are regulated sectors in countries where data sovereignty is important enough that the sector sticks to air-gapped on-prem hardware and does not use cloud services at all. They have the dollars to pay for more than what it would cost to run on the Cloud.
Interesting. So nobody would have had a problem with you running stuff on Chinese AI providers?
I have some inference I simply don't want to run on OAI, Anthropic, or Google because I don't want to run afoul of their "rules" and end up with a banned account, and this situation is only getting worse when it comes to doing fairly basic tasks like trying to secure your app against security problems.
Having worked in / adjacent several such industries, a lot of the question depends on scale.
A trillion-dollar business can easily trade dollars for the privacy. A business with $1M to spend won't even get a phone call with OpenAI or Anthropic, who were the only* previous players in town for doing this.
Worst-case example: Bootstrapped startup working in military.
It's also the case that an open model enables many more intermediate-cost solutions. E.g. providers certified for specific applications, on-prem rentals, etc.
* Omitting Azure, which gives some privacy for some $$$ on their models, but not at the level of high-security.
> Omitting Azure, which gives some privacy for some $$$ on their models, but not at the level of high-security.
If I were ranking third parties on their ability to safely handle my data without compromising it, I would rank Anthropic pretty low for things like Fable (where they more or less promise that they will misuse my data), but I want Azure pretty low in the sense that I fully expect them to be compromised.
I would tend to trust Amazon to avoid being compromised.
I’ve priced it out: max $135/month to run a dual Xeon 2U server with 3T RAM & 2x 22 core Xeon Gold. It’s the 2x 750W power supplies that ultimately determine opex. My power costs $0.124/kWh, the $135 assumes drawing maximum power continuously, and in that case, I can probably offset my heating bill a little bit in the winter, so maybe effectively a little bit lower.
I don’t know if that’s 100x more than I’d pay (opex-wise) with an nvidia setup, but I can say the one-time capex is a great deal cheaper. Avoiding VRAM and DDR5 (fast DDR4 should be OK) are the biggest cost savers. ECC RAM is worth the extra price. General datacenter-quality hardware has less price sensitivity, and plenty of bang for your buck.
Keep in mind that just because it has dual 750W power supplies that doesn't mean it's what its load will be, for a full CPU loaded wattage figure you'd need basically a pair of kill-a-watts plugged in inline on the feed for each poewr supply and then run stress-ng with artificial cpu stress on all cores for an hour.
Under heavy inference load you will find that the cpu usage is actually less as the bottleneck is the RAM bus speed. An older 2U rack server that is 600W load (typically a 1+1 power supply server when plugged into two kill-a-watt would show 300W on each, equal load balancing) when maxed out with stress-ng might be only 450W total running inference.
If you have 600kWh used in a month by running something 24x7 and your power is $0.15 a kWh, that's more like $90/mo (not counting cooling or any ancillary costs for the environment where it's in).
If you actually were running this thing at 80% or 100% load, then the first thing you'd want to is get a better PDU and then connect your servers to that (48V DC).
I dunno, K3 thinks a lot before it actually replies, and you might be in the ~1 tok/speed region or even "seconds / tokens", and with K3, you'd wait days if not weeks for a reply in that case.
Don't get me wrong, slow is sometimes better than "not at all", but depending on the performance, it might end up way too slow to even work for batched/async jobs like that.
I agree it's very likely to be painfully slow, I very much want to see some real world results from people who try it. Early testers will inform others on whether it's even worth trying. Results very much TBD right now. I don't have a system sitting around here with 2TB of greater of RAM that isn't already committed for other uses, regretfully.
Lets say an easy response takes 32k tokens in total, and to be generous, let's say it does 1 tok/s. This is already ~9 hours, and 32k reasoning tokens isn't even that much and as mentioned, K3 probably does the longest/most reasoning/thinking out of the available open weights models today, much like GLM. Just lowering that performance to 0.5 tok/s, would lead to ~18 hours for a simple prompt to receive an answer.
And then that's just for single prompts, what about agent harnesses, where before every tool call the model could reason a bunch?
I agree with you that real world results would be interesting, but I wouldn't hold my breath nor expect it to realistically be able to be useful. Still, people should try it, for science if nothing else :)
They are saying that AMD's new Epyc Venice CPU has 16 memory channels allowing up to 1.6Tb/s of bandwidth. Which is higher bandwidth than most non-HBM GPUs.
So full CPU local AI inference may become viable option in coming years.
It's a great concept but I think it would cross the line from 'very slow' to 'so slow it's unusable' at this size. Even if we say you have an NVME SSD that does 7GB/s reads, that's dramatically slower than being able to hold the whole thing in DRAM. Like the difference between 1.3 tok/s in RAM vs 0.1 tok/s with a colibri-like method.
edit: the results I have seen from people trying colibri with fast consumer grade PCI-E 4.0 NVME SSD are 0.1 tok/s on models that are <700B in size, things that are well under 800GB on disk. With something that's 3T in size it'll probably be a lot slower than hat.
For single stream inference of a MoE model, the size of active sparse parameters will matter a lot more than total parameters. This is generally around half of the reported active parameter count - the other half being a dense subset that can be easily cached in VRAM even on fairly modest consumer setups. So the achievable performance may be quite a bit better than a naïve assessment might suggest.
1536GB of DDR4 ECC server RAM is somewhere between $4000-6000 USD used right now, by the time you put in parallel enough NVME SSD to approach good speeds, you'd be approaching that (and also likely running out of PCI-E bus lanes directly attached to the same motherboard to reasonably do so).
Presumably it’s MoE and only needs to read a small fraction of the weights per token. Bonus points if you can get decent speculative decoding without becoming ALU-limited.
Speculative decoding is not really worthwhile for sparsely-loaded models. You end up paying in both memory bandwith and compute (loading experts based on wrongly-predicted tokens) which leaves you worse off overall. It becomes viable (even for sparse MoE) once you're batching so widely that you end up having to load most of your total weights anyway.
The performance bottleneck is not really so much the number of cores or processing power in each core, but the memory bus bandwidth to/from the CPU. I have an older dual socket xeon server here which is a CPU-only LLM test machine with 256GB of RAM and the actual CPU stress is not much, I can even quantify this by how little it spins up the CPU fans to meet thermal load (the CPUs are operating at nowhere near their 180W per socket max capacity, compared to like, crunching prime numbers or running cpuburn).
But the memory bus speed is fully committed when generating tokens or thinking.
Thats where the threadrippers really excelled. They had the lanes for memmory access. We might soon see the return of dinner plate-sized CPUs with thousands of pins.
Speaking of finetune, currently a common practice is LoRA over bnb 4-bit base model, but I think it's time to replace bnb with GGUF as the base model format. GGUF is actively supporting new model architectures and more aggressive quantizations.
I've made some proof of concept in https://github.com/woct0rdho/transformers5-qwen3.5-recipe . We can finetune Qwen3.5-35B-A3B in 16 GiB VRAM, and DeepSeek-V4-Flash (284B-A13B) in 90 GiB VRAM, without CPU offload. This works well on unified memory machines like Strix Halo.
Even so, larger models like Kimi-K3 still require multiple GPUs and nodes, and there are a lot more to do compare to single-GPU training.
> No, you don't. Without training cost you can infer only the marginal cost of serving this kind of models.
Still useful; "are the labs marginally profitable just on the marginal inference costs?" is still a useful question to answer. After all, if they aren't even profitable on inference in isolation, then we can expect to see large price increases.
If they are able to turn a marginal profit on inference alone, then perhaps the price increases won't be so severe (or perhaps they expand the time between generations so that they spend less on training but take longer to complete training).
"Are the labs profitable at all?" is, of course, a much more useful question, but that doesn't mean that the first question is completely useless.
We don't really know that, for OpenAI and Anthropic. We suspect that, but as far as I know, even they have stopped claiming that they are profitable on inference.
unless you think that Opus is 10T+ params, its pretty much impossible for inference not to be profitable when doing some basic napkin math on other open models, and if Kimi K3 is 3T params with the same performance as Opus then that means that China is actually way more technologically advanced than the American labs.
> If you get close in output quality, then does that matter?
When you're trying to estimate/infer the costs of serving the tokens and even include the cost of training the weights in order to output tokens then yeah, why wouldn't that matter?
Well, or if you're participating in a discussion on HN where the sub-topic happens to be "if labs are subsidising tokens on API pricing" and literally the cost of serving the tokens is relevant to the sub-topic people are trying to discuss...
Training cost is directly impacted by inference cost nowadays. Most of the gains come from RL these days, and that is highly dependant on inference (~7:1 inference:training in units of compute). That's mainly because you want many roll-outs for each training scenario.
Of course inference efficiency is dictated by model architecture, size, etc. You can still guesstimate some of those and have an idea about cost/serve at several size tiers.
I believe they are talking about the closed models' training costs.
I other words, the providers that will be offering K3 inference don't have any training costs to offset, so they are only charging for the inference itself. OAI/Anthropic would need to offset their R&D and training costs in order to not be selling API access at a loss.
Say a single Kimi K3 is deployed on 16 x B200s: how many concurrent users can that handle? I realize the question assumes a major simplification that everyone's prompts/sessions are the same.
>I realize the question assumes a major simplification that everyone's prompts/sessions are the same.
well, exactly.
that's tough to answer without just average sampling because some users will ask the model "what's todays date" or "what color is the sky?" and some users will ask "Let's rewrite the linux kernel in brainfuck."
Since the model is natively MXFP4, I think it'll be even more interesting on the hardware front. It'll comfortably fit on a 8x AMD MI355X node. I suspect that'll drive token prices down, further.
You're assuming inference providers are going to sell tokens at cost. You're also assuming that the inference providers have will optimized inference engine. I haven't seen that to be the case so far, to be honest.
Take a look at GLM 5 vs GLM 5.2 pricing -- GLM 5.2 cost more despite being the same model.
Take a look a look at DeepSeek, which hosts DS v4, profitably, yet others aren't able or willing to match the price.
Xioami's MiMo did match DS-V4's price, although we now know that DeepSeek set their pricing lower than they could have, due to the leaked memo, and simply decided to use "10 months to recover capex" as their yardstick. Interestingly "10 months to recover capex" is also the same price SpaceX is renting space to Anthropic and Google for.
I'll be honest, I typed that message while having morning coffee, so it's just a quick reaction from my part, not a heavily researched article in a journal :)
But I do think that the median price where this settles will tell us something about the floor at which it is profitable to serve this model.
> DeepSeek, which hosts DS v4, profitably
I specifically mentioned 3rd party providers, because there can be an argument that model creators themselves are subsidising tokens to gather training data for the next model. In fact, ds are public about their gathering of data (at least on openrouter they're marked as such). So that 0.x price point for dsv4-pro is likely subsidised.
I think it's unclear the the DS hosted prices are profitable. AFAIK that haven't claimed that.
OTOH, the multiple providers who have settled around the same price point ($3.48/M output tokens for multiple providers with good reputations) does indicate where it is profitable: https://openrouter.ai/deepseek/deepseek-v4-pro#providers
If it takes so much resource to run, how does the sharing of a single llm works? There is some interface that basically submits context/cache plus current promt, from each user, doing basically time-sharing compute?
If it is a mixture of experts (MoE) model like the 2.x models, won't this reduce the hardware needed to run the model?
The Kimi-K2.6 model is 1.1T parameters with 32B active parameters. With light quantization (Q6_K) that's enough to run it (slowly) on a single 5090. On a single B200 you can have 5-6 experts loaded into VRAM at a time. Realistically that would be 3-4 to account for the context. [!]
[!] With this and other MoE models it looks like an interesting area for research would be to detect or predict which models would be needed ahead of time. That way you could schedule the load into VRAM step before the weights are needed. That way you shouldn't lose much/any performance from offloading the weights to RAM.
You need whole weights in VRAM for optimal performance. Don't be confused by "experts" in the name -- you don't get to load static subset of experts and blast next 100 tokens with them. In typical MoE model they get switched "randomly" on every token, so all experts have to be readily available.
> With light quantization (Q6_K) that's enough to run it (slowly) on a single 5090.
Kimi K2.6 is released as INT4 already.
So 5090 with K2.6 is just gonna sit idle 99% of the time, waiting for next slice of weights to load.
5.6 Sol calculates that single 5090 in raw compute & memory bandwidth can run K2.6 at 35 t/s (256k context depth) -- if it somehow had enough memory to hold whole model in VRAM. Man, I hope HBF succeeds and Nvidia brings it to consumer cards in 5 years..
> In typical MoE model they get switched "randomly" on every token, so all experts have to be readily available.
It's worse than that: a typical MoE model routes a separate set of experts at every layer, not just every token! But in practice, RAM offload (for systems with non-unified VRAM) and even SSD offload still work surprisingly well given some amount of caching.
You can likely recover compute intensity and throughput by batching requests together, which (in practice, depending on sparsity) will end up reusing some of the loaded experts with high probability; though the obvious tradeoff is that having to store KV caches for the wider batches may leave you with less room to cache experts across layers and tokens.
(Plus if you're batching so widely that you end up loading essentially entire model layers, MTP then becomes applicable even for a MoE model. But this typically only applies if you're doing inference on a very large scale, or if your memory bandwidth is so scarce that you have to recover compute intensity by any means feasible.)
> The Kimi-K2.6 model is 1.1T parameters with 32B active parameters. With light quantization (Q6_K) that's enough to run it (slowly) on a single 5090
Without leveraging system RAM and/or SSDs, I don't think you can, or how exactly are you running this, if this is something you are doing today? With CPU/expert offloading you could probably do it with a 5090 + 1TB of RAM or something like that, but absolutely not on a single 5090 entirely within VRAM.
> There are a lot of optimisations that are not in the public sphere
Sure, but if we're participating in public discussions, isn't it more fun if we talk about things people can actually read and understand, rather than secret stuff other's can say work, but no can actually validate or know how it works?
It sounds like "hybrid approaches are much better than the public is aware, because everything else is private and secret", but also: ok, so what? No one can run that anyways, (yet?), so why it matters?
Alright, I guess I misunderstood. To be fair, this part:
> The Kimi-K2.6 model is 1.1T parameters with 32B active parameters. With light quantization (Q6_K) that's enough to run it (slowly) on a single 5090.
Is painting a very different perspective, even considering the latter parts it's hard to read that as "Of course offloading everything else that doesn't fit on the GPU itself". But anyways, it's been clarified now so no harm :)
I assume you mean putting only the 32B active parameters on the GPU, and the rest on a bunch of regular server DRAM like on a 768GB to 1024GB RAM server?
Because Kimi K2.6 in Q4 is about 584GB GGUF size on disk and will use slightly more than that in RAM, Q8 is 595GB.
You're talking about running this "at home" for 1 user, using a mix of VRAM and RAM (total should be ~1.5TB). That's certainly possible. It'll be slow, especially prompt processing, but doable for single users.
But my comment on running it was more towards serving this profitably at scale. You get much better throughput / unit of compute if you load everything in VRAM and serve many requests at the same time. That's how all inference providers are doing it.
I was talking about running this on a server, hence my comments re 1xB200. Obviously, the more hardware/VRAM you have the better/faster you can run these large models. But if you are a small/medium sized company you could feasibly do it on just one B200. It all depends on how much hardware you can afford to run.
This is a very insightful eye opening take. I haven't even thought of it this way. This really is the first open model to be as big as the frontier has been until now.
I think this release is actually great both ways when you think about it. We gonna be able to learn knowledge that labs have been hiding from us (e.g. cost like you mentioned). And labs could learn from whatever optimization techniques people come up with when trying to host this model.
It's honestly just good for everyone in my opinion.
We have a lossless compression codec (working on open sourcing it over the next couple of weeks) that reduces it down to its minimum entropy -- it cannot be compressed further. On all tested large models, it's a ratio of 1.34-1.23 -- and smaller models up to 3.76x. It also increases the effective bandwidth by the same rate.
Most likely, since they were acquired. But for us outsiders it would be a cool thing, to see if the delta is the same between kimi2.x + cursor data -> kimi3 + cursor data.
So nowadays the hardware and hosting providers must be in an optimization race, whoever can make the model just a bit smaller or more efficient (to fit on fewer/less powerful cards) will have a huge advantage and can make a lot of money.
I am curios what's the most profitable thing to "plant" (agriculture analogy) on the land (cards) that you have have: web hosting, vps, llms, image/video generation, etc
> SemiAnalysis estimates that Anthropic's current blended gross margin has risen to the mid-60% range, with the API business gross margin exceeding 80%
Of course, people will insist "they are lying", "why should we believe them, it's well known they subsidize API pricing", ...
Agreed. My (somewhat educated) guess is that top labs have healthy margins on API pricing. But this release will add another 3rd party / clear of conflict datapoint in this estimation.
Will the model even be competitive in 10 months though? Seems like models that reach top 20 on OpenRouter see 50% of all token spend by day 80, and 80% by day 180.
Or even the basis of the cost of hardware. There are lease deals, capacity traded for equity, various programs by Nvidia, there's absolutely massive depreciation, etc.
Until a few minutes ago there was a countdown page. (The weights haven't been released yet.) The countdown should be over in 19min, not sure why we're suddenly getting a 404.
I feel like most hardware to run LLMs on is shaped wrong for individuals.
It's either having a model struggling along with like 5-10 tokens per second on unified memory, or data center cards with hundreds of GB of VRAM consuming more than a kW of power. It doesn't seem like there's prosumer GPUs with like 180W-250W TDP and 128 GB or 256 GB of VRAM (one can dream). Then bifurcation and even just two of those cards would be kinda useful (albeit NVLink or equivalent would need to be commonplace).
Obviously nobody is running Kimi K3 locally without an insanely beefy homelab and lots of money to burn, but running GLM 5.2 would be cool at like ~100 tokens per second for a single session and maybe ~60 tokens per second with N subagents.
I have found that the "mostly didn't lose anything" Q8 large models that I want to run are all too large to run on the "only $3995!" 128GB max RAM systems that some people are buying, and definitely won't fit with any usable amount of context. Things like Qwen 3.5 122B Q8 or deepseek v4 flash Q8, or Laguna S 2.1 Q8 need 170-190GB of RAM including full context, which fits on a 256GB RAM dual socket workstation or rackmount server (sans GPU).
Copy and paste below from my notes and reported memory consumption with latest llama-server, assuming use of "--no-mmap" to load the entire thing into RAM at the time that llama-server launches.
DeepSeek-V4-Flash-UD-Q4_K_XL via unsloth
145GB on disk GGUF
0.03.323.204 I common_params_fit_impl: projected to use 178175 MiB of host memory
DeepSeek-V4-Flash-UD-Q8_K_XL via unsloth
151GB on disk GGUF
0.02.215.885 I common_params_fit_impl: projected to use 184636 MiB of host memory
Laguna-S-2.1-UD-Q8_K_X via unsloth
120GB on disk
0.01.616.119 I common_params_fit_impl: projected to use 172860 MiB of host memory
Qwen3.5-122B-A10B-UD-Q8_K_XL via unsloth
160GB on disk GGUF
165GB RAM use on launch, fresh context
0.04.976.905 I common_params_fit_impl: projected to use 170038 MiB of host memory
There's an emerging practice of using Q4 quants and Q8 KV cache for local inference.
At that point you can run both Qwen3.5-122B-A10B (my personal choice on Framework Desktop 128gb) and Laguna-S-2.1.
Now whether that's good enough for one's use-case remains to be determined. You can also get more out of those (local models and quantizations) if you further tweak the harness you use them with, but tbh this is where it gets too much work (at least for me and the time I have available).
> emerging practice of using Q4 quants and Q8 KV cache for local inference
That's not an emerging practice, it's a tested strategy that is these days only used as a last resort by those desperate to fit a model in memory. Some models do better than others, but generally the model quality suffers greatly under those conditions.
I have never seen anyone report "this produced really great results" from intentionally quantizing their context vs. leaving it at full precision which is the ordinary default.
LLM inference unfortunately also seems to be a task that's poorly formed for moderate consumer hardware,as a single user. For a single user use case, the load is bursty but requires the weights to be in memory already. So a multi user server that keeps the model weights in parts of its memory and then spends some more per user kv cache is wildly more efficient and the wildly expensive gpu cores aren't just sitting idle most of the time. Don't get me wrong, most desktop workloads are bursty, but the power needed to to them has gotten cheap enough that we can have way overkill for idle scenarios hardware just sitting on our desks.
Why does it have to be so bursty though? Just let it run multiple continuous-batched inferences overnight. This would work especially well in combination with SSD offload, and given any kind of sparse attention (common in more recent models) even swapping out the KV cache itself to disk might ultimately be a win. I wouldn't be surprised if something like that ultimately became feasible for single users running even K3 itself.
I find it difficult to always have one or more long-horizon tasks 'queued up' and ready to run... I find myself usually bottlenecked on design, review, or something similar that requires me being in the driver's seat. It's possible I could queue up a bunch of tasks, letting the LLM run off in multiple directions, but then I'd be less able to steer and course correct.
Just my experience though, I'm still figuring things out. Perhaps some subsets of tasks would be more ideal for these long-horizon workloads - exploration, multiple competing implementations, etc...
Use your favourite harness to help you find long-horizon tasks to have queued up. It's changed the structure of how my projects work a bit, and do you have to do some homework fast of how slow/fast your various providers or local inference are, but it's worth it. Start off with hobby projects so you get a feel for how it works.
I left something gargantuan running over the weekend (decompiling 1980s-era system software) and look forward to checking it out later today when I have a few free minutes.
In principle, a slower inference ought to be easier to steer and course-correct. You'd always be able to look at partial results, especially with a local model that doesn't hide its thinking.
it has to be so bursty for realtime usecases like chat, which is what most people are using it for today. of course, once (if) stuff like software dark factories start working out for the average person, then you'll be able to make full use of your hardware for workload where asynchonous execution is feasible and have it run several parallel tasks overnight, with an orchestrator managing the gpu(s) allocations.
Chat doesn't have to literally be realtime though, that's just the model most users have settled on. You could fire off your request, let it work unattended and check back on it later (perhaps after getting some notification from the chat frontend via RSS, Web Notifications API or similar that the full response is ready).
If your workload fits long batches throughout the night you could schedule them better, yeah. But I think very few have a usage pattern like that?
Given the hardware shortage in the world, I suspect renting ("sharing") via APIs will likely remain cheaper for the foreseeable future since each piece of hardware isn't sitting idle nearly as much.
I've been feeling for a while that as we keep increasing model size, we're going through the opposite of the PC revolution.
The "democratisation" talk from the frontier labs is especially egregious when they only release closed models (gpt-oss hardly counts) and are trying everything they can to make it harder to run open models.
Yeah, I've been experimenting with DiffusionGemma which sadly isn't as smart as Gemma itself, but holy hell is it FAST, and has image input as well, so doing things like "take a screenshot once per second + ask the model to categorize/model it WITH reasoning before" becomes realistic and doable.
I ended up implementing DiffusionGemma myself with Candle in Rust + CUDA, and it's quite literally the fastest model I've managed to run on my hardware.
The next mac Ultra will allow to run a big model locally with acceptable speed. But we need people to optimize it for that computer, and we’ll be more limited in models we can choose from
128GB is enough to run a large model, quantized, REAPed, with MoE and fast SSD for model weights
An AMD R9700 gets 20-50 TPS at ~300 watts on 27B. 100 TPS for the 35B MOE model. And there might be some more optimizations to that as AMD software support gets better with ROCm's latest versions.
I agree but worth noting that it's never gonna be very practical to run LLMs like this at home. Unless we have some sort of design breakthrough, the only "sensible" way to run them is at high batch levels on shared HW.
Like, yeah if I could spend a few grand on such a GPU I probably would coz I'm a rich nerd, but I'd acknowledge it as an extremely inefficient luxury, kinda like a sports car.
So I think you could say the real misfortune is that we don't really have the technology (be it computer tech or political/social tech) to do that shared-HW thing in way we can truly trust.
We could make LLM inference 100x cheaper to run at home efficiently, but that solution might need to be updated every 1-2 years, whereas current GPU are useful for various others tasks and last longer
You're right, and it's interesting to consider why. It's probably a combination of a few factors:
1) Local LLMs are a relatively new phenomenon and hardware takes years. Apple probably lucked into their unified memory architecture being suitable (in terms of memory size and bandwidth) for local LLMs, but it's only with the newest generations we're hearing about LLMs even being a consideration in their design process.
2) NVidia seem to be deliberately blocking consumers from taking this path - as evidenced by the removal of NVLink from the 30x0 series onwards - probably to protect their data center cards from internal competition?
3) Perhaps there's just not the market for it? It's feasible that the number of nerds interested local LLMs is very small in numbers, sales, and profit potential compared to gamers on the one side, and data centers on the other. (This would explain why AMD and Intel aren't trying to out-innovate NVidia in this area, despite it being an obvious opportunity.)
There will be a huge market for local inference once it's cheap and widely available.
Try to imagine output token speeds of 15,000 tok/s and a time-to-first-token of 200ms. (This has already been done for Llama 8B.)
Now imagine gargantuan context windows (2M, 4M, or even bigger); keep in mind the 1M context windows were science fiction a few years ago... now imagine having this on a local model on something like a phone or portable device that can be gathering data about things you're doing and constantly run inference for things useful to you. An obvious example of this would be a chatbot you can talk to that responds like a normal human conversation and doesn't have delays, but that's just scratching the surface.
The individual-shaped-hardware problem gets even sharper at the phone end. Shipping a 3B model on-device, the usable RAM budget after the OS and everything else is more like 2-4GBtotal, not per-model so it's not 'can I afford more VRAM', it's 'can I fit a language model and an STT model and embeddings without the OS killing my process'. Feels like phones are the most hardware-constrained 'individual' tier and get the least airtime in these kind of discussion. Is that because the models that fit are still too limited to be interesting, or something else?
Recently spent a few hours messing with bonsai 27B and ternary bonsai 27B at total weights + context fitting in slightly under 6GB RAM, and it's just dumb as hell. It writes what seems like grammatically correct content but it's extremely limited.
It will also happily hallucinate new names and content to fill in gaps in its knowledge, and present the hallucations in what looks like a correctly formatted sentence, so it could fool a person who doesn't know the subject matter. Like, I asked it for a description of Seattle and it hallucinated a name and description of a nonexistant tallest building in the city and suggested the view from its observation deck .
In no way was I surprised, it's asking a lot of under 6GB RAM usage. But I think for 99% of people they will get better results doing something over the network where the weights and inference engine are not on the device.
We already know that competition brought GLM 5.2 prices down roughly 45% since its release on June 16th (1.5 months ago), and the price downward slope is probably still going (I've been checking regularly and new providers keep fighting on price, I don't think prices have settled yet). For reference : https://openrouter.ai/z-ai/glm-5.2#providers
I saw arguments like "Providers cannot price less than their costs" in other comments. In economics, it's generally admitted that they shouldn't price less than their marginal costs, i.e. in their case roughly the cost of electricity, since a lot of these datacenters are not at capacity in terms of graphics cards usage (speculation since it's very easy to rent a GC for a couple hours on some providers). My guess is that someone will be selling tokens at less than electricity + depreciation of GCs soon, since there's a lot of competition and "smaller" data centers have overcapacity? This is speculation, correct me if I'm wrong
> My guess is that someone will be selling tokens at less than electricity + depreciation of GCs soon, since there's a lot of competition and "smaller" data centers have overcapacity? This is speculation, correct me if I'm wrong
My guess is they are selling you the tokens, then selling your tokens (data) onto someone else.
I see these conspiratorial arguments all the time and I think people massively overestimate the value of the average users tokens.
The problems with frontier models (design taste, ability to solve novel/difficult problems, etc) cannot be solved by throwing more slop from the average user at it.
Actually, most of the main deficiencies in current models stem from the fact that their data sets aren’t curated and specialized enough.
I don't think the goal of this data is necessarily model improvement.
I think it's marketing, advertising, and product refinement.
Ex: all the things Google wants your search data for.
It's somewhat silly to think the value of that data has changed much. Advertisers want to know what's popular and getting clicks and attention. Competitors want to know what features are getting used in their markets.
In the simplest case, think of this data as improving the harness, not the model.
Press x to doubt on the 45% number. The cheaper providers on open router are fp4 vs fp8 for official zai. There are some cheap fp8 ones (like novita) but the ui makes it seem like it's a temporary promotion, with their normal prices being almost equal to official zai (idk much about open router so not really sure what's going on with these discounts)
In my opinion, next step is to cut down on reasoning tokens while maintaining intelligence. The Chain of Thought and looping can still be an issue with these Chinese models. They in fact said K3 would improve in the area but it's still an issue that unfortunately harms the token cost wins a bit. OpenAI has been really impressive here, on the opposite end of this.
There is a really interesting startup in Prague that is doing just that. They fine-tuned Qwen 3.6 27b to have 46% fewer reasoning tokens while maintaining most of the performance characteristics. I'm interested to see if they continue down this path of optimizing reasoning for other models.
Given the frontier-level capabilities of Kimi K3, I'm wondering if it's possible to extract the core capabilities (fundamental reasoning and tool calling) of the model into a smaller one that consumer devices could run? Not sure exactly how, but either by heavy distillation or some other surgical method since Kimi has a Mixture of Experts architecture.
I think it's very valuable to have a smaller model that doesn't have any domain knowledge or facts built into its weights, but given the right context, could accurately reason about what to do and use the right tools.
I'm aware of colibri [1], but so far I've only seen extremely slow performance.
"I'd like a car that goes 300mph and gets 100mpg while doing it. I'm aware of a car that gets 100mpg but it is extremely slow."
You are describing fundamental tradeoffs. Getting more performance relative to model size and training token amount is what all of the labs are solving.
Labs are focusing on creating models, small or large, that perform well on various benchmarks, including general knowledge, domain-specific expertise, and agentic capabilities.
Asking for such a model while wanting to be small and fast would align with what you're describing, which I believe is different from what I'm pointing to.
The model I'm describing sacrifices domain knowledge and expertise for agentic reasoning and tool-calling capabilities at a reasonable speed.
This is probably one of the ways to achieve this. I see a plethora of such fine-tunes on HuggingFace [1], but they're either not much different than the base model or they're outright benchmaxxing.
There’s another way besides distillation that’s way cheaper: You can have the big model build prescriptive skills that the small model follows.
Take the “train” portion of tasks on some benchmark, have K3 complete it, and then output detailed descriptions of tools used and why, then run the validation tasks with some small model that has access to the skills.
K3 releasing as open source right at the same time that people are criticizing Opus for questionable performance (there's even a thread on HN about Opus 5's problems)... this is such a flex.
Given how fast models are improving, burning the weights into actual ROM is prohibitively expensive if you need (or want) to replace the chip every couple of months.
The alternative is on-TPU flash for storing the weights.
They release it before I could even get a kimi subscription because of waitlist? lol hard to believe that I might get a kimi subscription from a third party
I heard this is the talk in town these days. Why can't Meta keep up? With >10000000x more resources you'd think that they'd be able to introduce equally performant if not better open weight models
The SemiAnalysis piece on this is long but very much worth reading:
> The company appears burdened by far too many disparate groups that are over-optimizing for certain metrics as opposed to delivering usable technology for the company as a whole.
> And because Meta has a reputation for throwing money at problems and executing at high speed, these U-turns end up becoming more costly versus other companies that take a more disciplined or conservative approach. Suppliers also lose faith when given design wins are later cancelled. This has lead to less supply chain prioritization on new designs. Some suppliers favor focusing on Amazon or Google designs due to Meta’s frequent reshuffling.
> Few inside Meta’s chip division have a full understanding of why the company bought Rivos in the first place, and those who championed the deal internally have since gone quiet.
The cynic in me says maybe they would be further along if they hadn't spent $80 billion on trying to build the "Metaverse" VR world. I've never met anyone who actually uses it and to the best of my knowledge it has very low mass market uptake.
Because lack of talent and organizational disfunction matters a lot more than you think. The reason why OAI and Ant are always at the top is because of this and I’d say compute is third on the list.
I would argue that they actually don’t lack talent, they have an insane bench of really smart people. What they lack is any sort of direction and leadership. They are a ship lost in the ocean and up until now have been lucky to find a few treasures along the their way.
because imagine starting work every week and finding that your dumb ass CEO pivoted the company again and is ruining other peoples lives, and he then reorgs the management again so you now have your 5th leader this year.
Morale and momentum are huge things in companies, Zuck has been murdering both of those in Meta since... well naming it Meta.
You can make the same argument for closed models. Why spend hundreds of billions training larger and larger models when you can just use Chinese models? Spend that money somewhere else further up the stack where there’s more value. Let China do the training since they’re so efficient at it.
as a big tech company you have the resources to make many bets and do a lot of things at the same time. it's good to have some specialists with knowledge of model training "just in case".
As a completely one person, single sample anecdote, the 'heretic' uncensored Q8 GGUF variants several people have published of Qwen 3.5-122, 3.6-27B and 3.6-35B-A3B will very happily discuss just about any controversial topic that the CCP hates. Including lots of things that would get you thrown into prison if you published them in Mandarin on the domestic Chinese internet.
You could likely further de-censor a model by having a set of 'test' prompts in native Mandarin, Cantonese or really just about any other language. I don't speak any Chinese languages so I don't know if the published 'heretic' GGUF files some people have been throwing around will cooperate, or refuse, if you ask it in Mandarin for how to build a meth lab or precursors for semtex.
Indeed not, but I was saying there's more than ample precedent which is tested/working and actually doesn't refuse anything. Go to the Huggingface 'models' search interface and type in "heretic". Or uncensored. One example would be: https://huggingface.co/HauhauCS/Qwen3.5-122B-A10B-Uncensored...
Yeah, I think we will know more within a couple of days, once people actually download/run/test it. I'm sure the people adjacent to the 'heretic' developers will give it a test as soon as they get their hands on it. All very theoretical right now.
> I think we will know more within a couple of days
It's a 3T parameters model, with a weight format (MXFP4) still not completely integrated into the ecosystem, which only a few has the hardware to even do inference with, much less fine-tuning or more post-training. But sure, do sit and wait a few days :)
It would, indeed, be interesting to compare, given what we know about Anthropic’s censorship and political bias in their closed and more expensive models.
> For the first time, an open-weights LLM is right at the top.
Hmm, not quite true, I think that honor, for better or worse, goes to OpenAI. When they released GPT2 (or GPT1 for that matter) is was quite literally the SOTA in the ecosystem when it was released.
What China (and Japan) uses is YYYY年MM月DD日, which IMO is the superior date format since it's self-explanatory - the sections are spelled out right there!
Jumbling together year, month and day and having to separate them by counting digits is not an improvement for human readability. (And you have the year wrong.) Any of “27 July 2026”, “July 27, 2026” or “2026-07-27” would be superior.
It's good because it's an actual standard instead of 7/27 which is backwards for half the world. Besides, I'm actually from the future and they delayed the model a year, learning from Anthropic that good models are actually about to bring the end of society.
Don't even have to go that far, outside of white-collar jobs and some groups weirdly obsessed with scheduling, most people don't use calendars at all, but their manager/boss/significant-other does that for them :)
Not really. Spain's traditional format is dd/mm/yyyy with slashes. This applies for a good chunk of Europe. Germany/Austria uses dot, I think nordic countries embraced the dash. While you might see more adoption in offical/digital contexts, I just double checked a few popular spanish websites, all slashes.
Unfortunately I don't think modern consumer CPUs are physically capable of addressing enough RAM to even load the model into memory. We'd have to wait for some random person to make an extremely quantized version before we could reach those blazing speeds
Are they going to release Kimi K3.1? I’m eager to test it. According to rumors on X, it could outperform Fable. Could it be the first Chinese open-weight model to become the leading frontier model?
is there a realistic way to distill 2 consumer hardware friendly models with max ~200B and ~20B? Qwen did it, but would it be possible for 3rd parties (unsloth etc)?
Yeah, why not. Toughest part is running the hardware so you can create the traces for downstream training, but once over that hump, nothing would stop you from doing that no.
I believe they don't do torrents because it gives them a lot more control.
Like they can takedown or update downloads and they can prevent someone from trivially bypassing the license agreements you need to accept for some models.
I think it's shameful that Moonshot isn't providing us with party kits like Microsoft did with the Windows 7 Launch Party kit. How am I supposed to properly celebrate this without fun Kimi-themed quizzes for my guests?
I think the results might be underwhelming - AI providers need to turn a profit and can't subsidize, and they're working off of the commodity hardware everyone does.
I wouldn't be surprised if they started offering potentiall bad quantizations with much reduced capability at lower prices (without telling the users, of course)
I would be surprised, considering that OpenRouter requires disclosing the quantization and shows automatic benchmarks to compare between providers for the same model.
As long as they are transparent about what quant they serve the model and any other optimization they do that also affects performance of inferred tokens.
There's no going back on this. This is putting a very capable intelligence in the hands of the masses. Private companies in the US are aching for Trump's protectionism but it'll do nothing. The hardware needed to run this is ofc prohibitive, but actually putting it out there feels like a 'RSA source code on t-shirt' moment for humanity.
No, luckily private companies in the US are aching for this to NOT be banned. NVIDIA, Microsoft, etc. just released that letter. We’re saved from the trillionaire companies (OpenAI, Anthropic) by the other trillionaire companies acting in self-interest (hosting and hardware).
Strange communists, giving away such an expensive model to the public.
On the other note, can't wait to see 1bit quantisation soon and how it performs in benchmarks, if it performs really well in benchmarks, would be very good news for GPU hosting providers, to offer "Opus 4.5 level model at the cost of Haiku 4.5"
The short answer is no, it won't work on your home computer. In it's current form it needs something like 594 GB of memory, far outside what you can reasonably run on normal consumer hardware in 2026.
If you have really high end hardware, you might be able to squeeze a heavily quantized version of Kimi-K3 onto your rig, but it will be too slow or too lobotomized to be useful.
This does put a near state-of-the-art open weights model within reach of what a small or medium business could afford if there's a case for local inference. It's probably not as good as Claude Fable or ChatGPT Sol. But if you're an organization that has a genuine need to run inference locally, this is a real possibility.
Is this for your homelab? Not in any practical sense.
Is this a possibility for organizations that can justify $1M or so on hardware for a near SOTA model they have full control over? Yeah, absolutely.
The full K3 model will probably be way more than 594GB, that's more of a plausible range for Kimi 2.x. You'll probably be able to test run this model at full or near-full precision using SSD offload, but only at very slow speeds - probably slow enough that you'll be forced to let inferences run overnight or even spanning multiple days. Mind you, that's still useful enough for many casual users, given that they're running a near-SOTA model!
FYI huggingface refers to the alien from the Aliens movies that we need to prevent from reaching earth at any cost because it means the end of civilization.
Just checking in because y'all sound good with that.
This will be interesting for a few reasons. First, depending on where the median pricing settles w/ 3rd party providers will tell us what it costs to serve a 3T model. Since it's going to be mxfp4 native, it'll take ~1.5TB of VRAM to host this, which is juuust at the limit of 8xb200s (but realistically you'll need 16x for context / throughput optimisation). Won't be cheap to host, but at least we should get some range of $/MTok for a 3T model. Then we'll be able to guesstimate if "labs are subsidising tokens on API pricing".
Also interesting to see what effort it will take to fine-tune this beast. The latest AISI benchmarks on cybersec place it above glm5.2, but still way way behind SotA closed models. Some fine-tuning might be needed here. Also, interesting to see if Cursor does another training round on it, to directly compare it w/ kimi2.6/2.7 fine-tunes (composer series) and grok4.5.
Also also, interesting to see if someone takes on distilling (proper distillation, w/ training the entire distribution) from this into smaller models. (dsv4-kimi should be really good, since dsv4 is very cheap to serve)
It will be very interesting to see what kind of 'slow' performance people get from running it on a no GPU, but tons of RAM server (like a dual or quad socket xeon with 1.5 to 3TB of RAM). For the purpose of giving it longer duration tasks to generate a piece of something and come back and check on what it has done in 4 or 6 hours. Even if the output is like 5-6 tok/s, that might be usable for some purposes.
Huge price difference in what you can do with buying a used 4U rackmount server and putting 3TB of RAM in it (64GB DIMMs x quantity 32 in a quad socket xeon, you can see some benchmark prices on eBay for sets of 16 or 32 matched 64GB ECC DIMMs) for <$30,000, vs the cost of trying to run it on real GPU hardware.
Now obviously, as of the time I write this, the full precision hasn't been released nor has anyone like unsloth run it through quantization yet to produce a "Q8" or "Q8-XL" variant of it. But I think it's going to need more than 1536GB of RAM, with a usable and large amount of context, more like 2TB and preferably 2.5 to 3TB.
I also predict that people who try to run it in Q4 and Q6 will get the worst of both worlds, less precision/lost knowledge but also not reliable output that comes out too slow. In my personal opinion if I'm going to deal with something that is smart but slow and running on limited budget hardware, I need it to be Q8.
> Even if the output is like 5-6 tok/s, that might be usable for some purposes.
You'll spend ~100x more on electricity than the API cost to have it run on someone else's GPU at several hundred tokens per second.
I think some sort of extreme data privacy requirement is the only situation that justifies this, but the intersection of {needs absolute data privacy, needs to run SOTA model, cannot afford GPUs} is really really narrow. I wouldn't be surprised if this is an empty set.
There are a number of use cases where sending the contents of your context and prompts (and the resulting output) to a 3rd party service is off the table as an option, and people will compromise speed for data sovereignty. And not everyone's electricity is equally expensive, I pay about $0.075 USD per kWh. It would for example cost me about $48 a month of electricity (not counting cost of cooling) to run a quad socket Dell R940 for a month.
That's an unusually low electric rate for the US - way below the lowest state average which is Idaho at 12.4 cents. It's certainly possible that you are getting 7.5 cents including delivery, but I've had friends say that they're "getting 13 cents per kWh" here in Massachusetts, but that's just the supply rate and the delivery is another ~18 cents.
There are parts of states like Grant County Washington that have cheap hydro power, but it's very rare for power to be that cheap in the US. Even if this applies to you, it won't apply to the vast majority of people on here who will have electric rates 2-4x higher.
Average electric rates by region:
https://www.eia.gov/electricity/monthly/epm_table_grapher.ph...They are most likely not based in the US, but converting to USD to make comparison easier.
Great, so the other member of the set matters for you more than cost.
Do you actually need to run the state of art model at 5 tokens per second instead of a qwen or whatever 7b or 30b model at 100 tokens per second?
>Do you actually need to run the state of art model at 5 tokens per second instead of a qwen or whatever 7b or 30b model at 100 tokens per second?
Some people like doing things they want to do. Do I actually need to buy expensive pigments from europe to make paintings of flowers? My camera produces a much more accurate representation.
Very good description of it. It does seem like a bit of a rhetorical question to ask a forum that has a very high population of Linux and BSD users why they might desire to have the option to do something themselves rather than relying on an external packaged ready to go product.
Do I really need to? No, not really. The 27B full density, 35B MoE, 70B and 122B models I have in use get me 95% of the way there on a lot of things. Particularly when dealing with languages and systems where I have at least an intermediate level of knowledge on, to know whether something is going down a dead end, using a wrong method, metaphorically chasing its tail, or is producing valid output.
On the other hand, would it be cool to also have a really big thing as an ancillary tool that I could throw a request into opencode before going to bed, let it crank away and take a look at what it's done 7 hours later? Yeah, particularly if I (very much an unknown quantity at this time) could be confident that it builds high quality, syntax valid, appropriately commented and not absurd code.
As someone who has worked in two industries that are at the maximal end of data sensitivity and privacy this comes across as a tinfoil hat issue not a real business requirement. In such cases we've always found ways to trade dollars for the privacy we need without having to run our own inference at excruciating slow speeds.
Do you mean by trading dollars for the privacy you need as:
a) Contracting with a third-party independent inference provider who will run your choice of model on fast hardware that they own, with all appropriate data security/privacy/contractual/compliance protection in place
or
b) Contracting with the original creators of the model to run inference via their API and with assurances that all the same data protection is in place
or
c) Spending the money to buy your own inference hardware to run it on something you fully own/control at proper usable speeds?
Edit: Everything I've been writing in this thread is mostly within the context of being able to evaluate K3 and its usefulness to be self-hosted as a preliminary proof of concept or test of feasibility of a new thing, such as on <$20,000 of server hardware, before proceeding to spend 300-400k on GPU-related hardware, or external third party services/ongoing billing.
A) is very doable with e.g. Amazon Bedrock.
They'll give you HIPAA compliance, they even have a data center for US government classified data, they can give you European data sovereignty. And with OpenAI and Anthropic models to boot, you don't even have to settle for open weights.
What kind of privacy needs do you really have beyond that?
It is not my use case but given recent political developments in international relations caused by the executive branch of the US government, off the top of my head, I could think of a lot of European or Canadian firms for which that would not be an option. No matter what they might promise about European sovereignty. For a good 'ol patriotic US domestic company? Sure.
Yes, its the US cloud act risk EU companies run up against on hyperscalers like MS/AWS.
Even for EU companies running open weights on EU stacks LLM inference on the GPU must process plaintext and I can't find any EU provider with NVIDIA H100/H200/Blackwell CC mode plus SEV-SNP or TDX, where you can cryptographically verify the workload ran somewhere the operator cannot inspect.
Personal compute is therefore the only option if you want personal autonomy privacy for IP &c. Maybe another option is to use cloud compute rented to fine tune a personal model that suits your own needs that would help bring the cost down, I don't know enough about this area to know if it kills the "intelligence" of those domains due to limited ?cross-verification within the LLM.
There are regulated sectors in countries where data sovereignty is important enough that the sector sticks to air-gapped on-prem hardware and does not use cloud services at all. They have the dollars to pay for more than what it would cost to run on the Cloud.
Interesting. So nobody would have had a problem with you running stuff on Chinese AI providers?
I have some inference I simply don't want to run on OAI, Anthropic, or Google because I don't want to run afoul of their "rules" and end up with a banned account, and this situation is only getting worse when it comes to doing fairly basic tasks like trying to secure your app against security problems.
Having worked in / adjacent several such industries, a lot of the question depends on scale.
A trillion-dollar business can easily trade dollars for the privacy. A business with $1M to spend won't even get a phone call with OpenAI or Anthropic, who were the only* previous players in town for doing this.
Worst-case example: Bootstrapped startup working in military.
It's also the case that an open model enables many more intermediate-cost solutions. E.g. providers certified for specific applications, on-prem rentals, etc.
* Omitting Azure, which gives some privacy for some $$$ on their models, but not at the level of high-security.
> Omitting Azure, which gives some privacy for some $$$ on their models, but not at the level of high-security.
If I were ranking third parties on their ability to safely handle my data without compromising it, I would rank Anthropic pretty low for things like Fable (where they more or less promise that they will misuse my data), but I want Azure pretty low in the sense that I fully expect them to be compromised.
I would tend to trust Amazon to avoid being compromised.
I’ve priced it out: max $135/month to run a dual Xeon 2U server with 3T RAM & 2x 22 core Xeon Gold. It’s the 2x 750W power supplies that ultimately determine opex. My power costs $0.124/kWh, the $135 assumes drawing maximum power continuously, and in that case, I can probably offset my heating bill a little bit in the winter, so maybe effectively a little bit lower.
I don’t know if that’s 100x more than I’d pay (opex-wise) with an nvidia setup, but I can say the one-time capex is a great deal cheaper. Avoiding VRAM and DDR5 (fast DDR4 should be OK) are the biggest cost savers. ECC RAM is worth the extra price. General datacenter-quality hardware has less price sensitivity, and plenty of bang for your buck.
Keep in mind that just because it has dual 750W power supplies that doesn't mean it's what its load will be, for a full CPU loaded wattage figure you'd need basically a pair of kill-a-watts plugged in inline on the feed for each poewr supply and then run stress-ng with artificial cpu stress on all cores for an hour.
Under heavy inference load you will find that the cpu usage is actually less as the bottleneck is the RAM bus speed. An older 2U rack server that is 600W load (typically a 1+1 power supply server when plugged into two kill-a-watt would show 300W on each, equal load balancing) when maxed out with stress-ng might be only 450W total running inference.
If you have 600kWh used in a month by running something 24x7 and your power is $0.15 a kWh, that's more like $90/mo (not counting cooling or any ancillary costs for the environment where it's in).
If you actually were running this thing at 80% or 100% load, then the first thing you'd want to is get a better PDU and then connect your servers to that (48V DC).
(Context: Parent comment was edited after I wrote this comment)
Where in the world are you finding that much RAM in a racked server for $200/month?
I think he means electrical bill at his estimated wattage load of the server and his known kWh cost, not rented server/hosting cost.
Aha, right. That makes a lot more sense.
At my house. I have 5Gbps fiber and could pay for 10 or 25 if I need it.
Gotcha. But to be clear, you’re talking only about energy usage, correct?
Yes, what other opex is there? It will have good ventilation, I’m not worried about cooling.
I dunno, K3 thinks a lot before it actually replies, and you might be in the ~1 tok/speed region or even "seconds / tokens", and with K3, you'd wait days if not weeks for a reply in that case.
Don't get me wrong, slow is sometimes better than "not at all", but depending on the performance, it might end up way too slow to even work for batched/async jobs like that.
I agree it's very likely to be painfully slow, I very much want to see some real world results from people who try it. Early testers will inform others on whether it's even worth trying. Results very much TBD right now. I don't have a system sitting around here with 2TB of greater of RAM that isn't already committed for other uses, regretfully.
Lets say an easy response takes 32k tokens in total, and to be generous, let's say it does 1 tok/s. This is already ~9 hours, and 32k reasoning tokens isn't even that much and as mentioned, K3 probably does the longest/most reasoning/thinking out of the available open weights models today, much like GLM. Just lowering that performance to 0.5 tok/s, would lead to ~18 hours for a simple prompt to receive an answer.
And then that's just for single prompts, what about agent harnesses, where before every tool call the model could reason a bunch?
I agree with you that real world results would be interesting, but I wouldn't hold my breath nor expect it to realistically be able to be useful. Still, people should try it, for science if nothing else :)
You can rent one in the cloud to try it
They are saying that AMD's new Epyc Venice CPU has 16 memory channels allowing up to 1.6Tb/s of bandwidth. Which is higher bandwidth than most non-HBM GPUs.
So full CPU local AI inference may become viable option in coming years.
> But I think it's going to need more than 1536GB of RAM, with a usable and large amount of context, more like 2TB and preferably 2.5 to 3TB.
The model is known to be MXFP4 according to Kimi's release blog post, so the model weights will be less than 1536GB: https://www.kimi.com/blog/kimi-k3
Also, their previous models were native INT4, so it would be weird if they went larger now.
> running it on a no GPU, but tons of RAM server
Or from SSD using something like Colibri[1]. Not going to be quick, but at least runable.
[1]: https://github.com/JustVugg/colibri
It's a great concept but I think it would cross the line from 'very slow' to 'so slow it's unusable' at this size. Even if we say you have an NVME SSD that does 7GB/s reads, that's dramatically slower than being able to hold the whole thing in DRAM. Like the difference between 1.3 tok/s in RAM vs 0.1 tok/s with a colibri-like method.
edit: the results I have seen from people trying colibri with fast consumer grade PCI-E 4.0 NVME SSD are 0.1 tok/s on models that are <700B in size, things that are well under 800GB on disk. With something that's 3T in size it'll probably be a lot slower than hat.
For single stream inference of a MoE model, the size of active sparse parameters will matter a lot more than total parameters. This is generally around half of the reported active parameter count - the other half being a dense subset that can be easily cached in VRAM even on fairly modest consumer setups. So the achievable performance may be quite a bit better than a naïve assessment might suggest.
On a server machine you can have more than 100GB/s of NVMe if you parallelize (RAID 0 and the like). But it's still gonna be noticeably slower.
1536GB of DDR4 ECC server RAM is somewhere between $4000-6000 USD used right now, by the time you put in parallel enough NVME SSD to approach good speeds, you'd be approaching that (and also likely running out of PCI-E bus lanes directly attached to the same motherboard to reasonably do so).
It claims to support using multiple devices RAID-0 style, which should boost performance, but yea probably not very useful for most.
But still fun you can run it at home.
> Even if the output is like 5-6 tok/s
On a 3T model I’d imagine you’d be closer to 0.05 tks
Presumably it’s MoE and only needs to read a small fraction of the weights per token. Bonus points if you can get decent speculative decoding without becoming ALU-limited.
Speculative decoding is not really worthwhile for sparsely-loaded models. You end up paying in both memory bandwith and compute (loading experts based on wrongly-predicted tokens) which leaves you worse off overall. It becomes viable (even for sparse MoE) once you're batching so widely that you end up having to load most of your total weights anyway.
> Speculative decoding is not really worthwhile for sparsely-loaded models.
If wonder if you can train a model to optimize this, by trying to make the expert selection sticky across a few tokens, without too much quality loss.
Another fun idea might be to try to build a model where the router chooses the expert 1-3 tokens in advance.
Those old LTT videos of high core-count threadrippers running GPU benchmarks become more relevant each day.
The performance bottleneck is not really so much the number of cores or processing power in each core, but the memory bus bandwidth to/from the CPU. I have an older dual socket xeon server here which is a CPU-only LLM test machine with 256GB of RAM and the actual CPU stress is not much, I can even quantify this by how little it spins up the CPU fans to meet thermal load (the CPUs are operating at nowhere near their 180W per socket max capacity, compared to like, crunching prime numbers or running cpuburn).
But the memory bus speed is fully committed when generating tokens or thinking.
Thats where the threadrippers really excelled. They had the lanes for memmory access. We might soon see the return of dinner plate-sized CPUs with thousands of pins.
The epyc Venice SP7 socket is apparently 9324 pins
https://x.com/tomshardware/status/2066846693778510331
We are going to need a bigger boat.
https://www.cerebras.ai/
Speaking of finetune, currently a common practice is LoRA over bnb 4-bit base model, but I think it's time to replace bnb with GGUF as the base model format. GGUF is actively supporting new model architectures and more aggressive quantizations.
I've made some proof of concept in https://github.com/woct0rdho/transformers5-qwen3.5-recipe . We can finetune Qwen3.5-35B-A3B in 16 GiB VRAM, and DeepSeek-V4-Flash (284B-A13B) in 90 GiB VRAM, without CPU offload. This works well on unified memory machines like Strix Halo.
Even so, larger models like Kimi-K3 still require multiple GPUs and nodes, and there are a lot more to do compare to single-GPU training.
I dont quite understand why GGUF is better optimized. Are the performances better for the same amount of VRAM ?
I cannot be sure what the likes of Cursor have done, but I think it's incredibly unlikely that they have trained a QLoRA for Composer.
It's almost certainly full parameter post training of the original model weights.
> Then we'll be able to guesstimate if "labs are subsidising tokens on API pricing".
No, you don't. Without training cost you can infer only the marginal cost of serving this kind of models.
Moreover, you don't know the actual size of closed models (what if Fable is a 10T model? What if it's 1T?)
> No, you don't. Without training cost you can infer only the marginal cost of serving this kind of models.
Still useful; "are the labs marginally profitable just on the marginal inference costs?" is still a useful question to answer. After all, if they aren't even profitable on inference in isolation, then we can expect to see large price increases.
If they are able to turn a marginal profit on inference alone, then perhaps the price increases won't be so severe (or perhaps they expand the time between generations so that they spend less on training but take longer to complete training).
"Are the labs profitable at all?" is, of course, a much more useful question, but that doesn't mean that the first question is completely useless.
I wasn't saying it isn't useful, I was saying that you cannot infer even the marginal cost of closed models.
The K3 maths can turn true only if the models size is roughly the same and the labs didn't find any better way to run inference at scale.
We know labs make money on inference, and we know they lose a lot of money on inference+training.
> We know labs make money on inference
We don't really know that, for OpenAI and Anthropic. We suspect that, but as far as I know, even they have stopped claiming that they are profitable on inference.
unless you think that Opus is 10T+ params, its pretty much impossible for inference not to be profitable when doing some basic napkin math on other open models, and if Kimi K3 is 3T params with the same performance as Opus then that means that China is actually way more technologically advanced than the American labs.
So which is it?
Just out of curiosity, based on what we know for sure they(OAI+A) make money on pure inference and lose on inference+training?
OpenAI's financials leaked and showed this pretty convincingly.
Anthropic was probably profitable last quarter, without training costs: https://www.wsj.com/tech/ai/mind-blowing-growth-is-about-to-...
> Without training cost you can infer only the marginal cost of serving this kind of models.
Which is by far the most interesting number of the two.
> Moreover, you don't know the actual size of closed models (what if Fable is a 10T model? What if it's 1T?)
If you get close in output quality, then does that matter?
> If you get close in output quality, then does that matter?
When you're trying to estimate/infer the costs of serving the tokens and even include the cost of training the weights in order to output tokens then yeah, why wouldn't that matter?
That only matters if you are an investor not if you are a consumer.
Well, or if you're participating in a discussion on HN where the sub-topic happens to be "if labs are subsidising tokens on API pricing" and literally the cost of serving the tokens is relevant to the sub-topic people are trying to discuss...
Consumers want better models too, of course it matters
> Which is by far the most interesting number of the two.
Only if you don't have to continuously train new models, and you are not at a runway risk.
Training cost is directly impacted by inference cost nowadays. Most of the gains come from RL these days, and that is highly dependant on inference (~7:1 inference:training in units of compute). That's mainly because you want many roll-outs for each training scenario.
Of course inference efficiency is dictated by model architecture, size, etc. You can still guesstimate some of those and have an idea about cost/serve at several size tiers.
That isn't really relevant to GP's point. We still don't know what training costs because we don't know how much RL is done.
Gross margins are insanely important, possibly the most important single metric if for some reason you were forced to choose one.
Only if you assume that at some point, for any reason, there will be "the model" that doesn't need costly retraining.
I guess this is one of the reason Anthropic i so "active" for asking for a development break.
"I want my competitors to stop competing" is an interesting ask for someone who's currently charging the highest prices in the entire industry.
>No, you don't. Without training cost you can infer only the marginal cost of serving this kind of models.
Are you talking about Kimi's training cost or the training cost of the model(s) that Kimi distilled?
Because Moonshot didn't even incur the majority of the training costs either
> the training cost of the model(s) that Kimi distilled?
The distillation process involves getting conversation traces from the model you are distilling from, and then training your model against them.
You still have to do the training!
I believe they are talking about the closed models' training costs.
I other words, the providers that will be offering K3 inference don't have any training costs to offset, so they are only charging for the inference itself. OAI/Anthropic would need to offset their R&D and training costs in order to not be selling API access at a loss.
Say a single Kimi K3 is deployed on 16 x B200s: how many concurrent users can that handle? I realize the question assumes a major simplification that everyone's prompts/sessions are the same.
>I realize the question assumes a major simplification that everyone's prompts/sessions are the same.
well, exactly.
that's tough to answer without just average sampling because some users will ask the model "what's todays date" or "what color is the sky?" and some users will ask "Let's rewrite the linux kernel in brainfuck."
Since the model is natively MXFP4, I think it'll be even more interesting on the hardware front. It'll comfortably fit on a 8x AMD MI355X node. I suspect that'll drive token prices down, further.
You're assuming inference providers are going to sell tokens at cost. You're also assuming that the inference providers have will optimized inference engine. I haven't seen that to be the case so far, to be honest.
Take a look at GLM 5 vs GLM 5.2 pricing -- GLM 5.2 cost more despite being the same model.
Take a look a look at DeepSeek, which hosts DS v4, profitably, yet others aren't able or willing to match the price.
Xioami's MiMo did match DS-V4's price, although we now know that DeepSeek set their pricing lower than they could have, due to the leaked memo, and simply decided to use "10 months to recover capex" as their yardstick. Interestingly "10 months to recover capex" is also the same price SpaceX is renting space to Anthropic and Google for.
I'll be honest, I typed that message while having morning coffee, so it's just a quick reaction from my part, not a heavily researched article in a journal :)
But I do think that the median price where this settles will tell us something about the floor at which it is profitable to serve this model.
> DeepSeek, which hosts DS v4, profitably
I specifically mentioned 3rd party providers, because there can be an argument that model creators themselves are subsidising tokens to gather training data for the next model. In fact, ds are public about their gathering of data (at least on openrouter they're marked as such). So that 0.x price point for dsv4-pro is likely subsidised.
I think it's unclear the the DS hosted prices are profitable. AFAIK that haven't claimed that.
OTOH, the multiple providers who have settled around the same price point ($3.48/M output tokens for multiple providers with good reputations) does indicate where it is profitable: https://openrouter.ai/deepseek/deepseek-v4-pro#providers
If it takes so much resource to run, how does the sharing of a single llm works? There is some interface that basically submits context/cache plus current promt, from each user, doing basically time-sharing compute?
I believe they batch requests so the same weights in vram are shared across many requests https://www.baseten.co/blog/continuous-vs-dynamic-batching-f...
If it is a mixture of experts (MoE) model like the 2.x models, won't this reduce the hardware needed to run the model?
The Kimi-K2.6 model is 1.1T parameters with 32B active parameters. With light quantization (Q6_K) that's enough to run it (slowly) on a single 5090. On a single B200 you can have 5-6 experts loaded into VRAM at a time. Realistically that would be 3-4 to account for the context. [!]
[!] With this and other MoE models it looks like an interesting area for research would be to detect or predict which models would be needed ahead of time. That way you could schedule the load into VRAM step before the weights are needed. That way you shouldn't lose much/any performance from offloading the weights to RAM.
You need whole weights in VRAM for optimal performance. Don't be confused by "experts" in the name -- you don't get to load static subset of experts and blast next 100 tokens with them. In typical MoE model they get switched "randomly" on every token, so all experts have to be readily available.
> With light quantization (Q6_K) that's enough to run it (slowly) on a single 5090. Kimi K2.6 is released as INT4 already.
So 5090 with K2.6 is just gonna sit idle 99% of the time, waiting for next slice of weights to load.
5.6 Sol calculates that single 5090 in raw compute & memory bandwidth can run K2.6 at 35 t/s (256k context depth) -- if it somehow had enough memory to hold whole model in VRAM. Man, I hope HBF succeeds and Nvidia brings it to consumer cards in 5 years..
> In typical MoE model they get switched "randomly" on every token, so all experts have to be readily available.
It's worse than that: a typical MoE model routes a separate set of experts at every layer, not just every token! But in practice, RAM offload (for systems with non-unified VRAM) and even SSD offload still work surprisingly well given some amount of caching.
You can likely recover compute intensity and throughput by batching requests together, which (in practice, depending on sparsity) will end up reusing some of the loaded experts with high probability; though the obvious tradeoff is that having to store KV caches for the wider batches may leave you with less room to cache experts across layers and tokens.
(Plus if you're batching so widely that you end up loading essentially entire model layers, MTP then becomes applicable even for a MoE model. But this typically only applies if you're doing inference on a very large scale, or if your memory bandwidth is so scarce that you have to recover compute intensity by any means feasible.)
> The Kimi-K2.6 model is 1.1T parameters with 32B active parameters. With light quantization (Q6_K) that's enough to run it (slowly) on a single 5090
Without leveraging system RAM and/or SSDs, I don't think you can, or how exactly are you running this, if this is something you are doing today? With CPU/expert offloading you could probably do it with a 5090 + 1TB of RAM or something like that, but absolutely not on a single 5090 entirely within VRAM.
Yes hybrid approaches are much better than people realise.
There are a lot of optimisations that are not in the public sphere, source working on start up in this space
> There are a lot of optimisations that are not in the public sphere
Sure, but if we're participating in public discussions, isn't it more fun if we talk about things people can actually read and understand, rather than secret stuff other's can say work, but no can actually validate or know how it works?
It sounds like "hybrid approaches are much better than the public is aware, because everything else is private and secret", but also: ok, so what? No one can run that anyways, (yet?), so why it matters?
Yes, that's what I was saying w.r.t. expert offloading, i.e. ensuring that the GPU could fit the active parameters not all the parameters.
Alright, I guess I misunderstood. To be fair, this part:
> The Kimi-K2.6 model is 1.1T parameters with 32B active parameters. With light quantization (Q6_K) that's enough to run it (slowly) on a single 5090.
Is painting a very different perspective, even considering the latter parts it's hard to read that as "Of course offloading everything else that doesn't fit on the GPU itself". But anyways, it's been clarified now so no harm :)
I assume you mean putting only the 32B active parameters on the GPU, and the rest on a bunch of regular server DRAM like on a 768GB to 1024GB RAM server?
Because Kimi K2.6 in Q4 is about 584GB GGUF size on disk and will use slightly more than that in RAM, Q8 is 595GB.
https://huggingface.co/unsloth/Kimi-K2.6-GGUF
You're talking about running this "at home" for 1 user, using a mix of VRAM and RAM (total should be ~1.5TB). That's certainly possible. It'll be slow, especially prompt processing, but doable for single users.
But my comment on running it was more towards serving this profitably at scale. You get much better throughput / unit of compute if you load everything in VRAM and serve many requests at the same time. That's how all inference providers are doing it.
I was talking about running this on a server, hence my comments re 1xB200. Obviously, the more hardware/VRAM you have the better/faster you can run these large models. But if you are a small/medium sized company you could feasibly do it on just one B200. It all depends on how much hardware you can afford to run.
This is a very insightful eye opening take. I haven't even thought of it this way. This really is the first open model to be as big as the frontier has been until now.
I think this release is actually great both ways when you think about it. We gonna be able to learn knowledge that labs have been hiding from us (e.g. cost like you mentioned). And labs could learn from whatever optimization techniques people come up with when trying to host this model.
It's honestly just good for everyone in my opinion.
We have a lossless compression codec (working on open sourcing it over the next couple of weeks) that reduces it down to its minimum entropy -- it cannot be compressed further. On all tested large models, it's a ratio of 1.34-1.23 -- and smaller models up to 3.76x. It also increases the effective bandwidth by the same rate.
Exploring compression algorithms for weights is a good idea, and I hope you have a successful product. However, if you can prove this statement:
> reduces it down to its minimum entropy -- it cannot be compressed further.
I think you could make a lot more money elsewhere :-)
https://en.wikipedia.org/wiki/Kolmogorov_complexity#Formal_p...
We're not an AI company ... nor do we have any reason to use it. Just a fun idea that was fruitful.
That's very interesting. Does that mean you can reduce say, a 30B class Q8 from ~30 GB down to 10 GB or less?
704gb -> 564gb; 358 gb -> 270 gb; 28.79 gb -> 7.65 gb; 439 gb -> 93 gb
It depends on the total entropy of the model. Smaller models have less entropy.
pardon my ignorance but is fine tuning still considered viable in face of rapid model releases. Is it really worth it?
my friends whove tried in their companies gave up on it.
> The latest AISI benchmarks on cybersec place it above glm5.2, but still way way behind SotA closed models.
Sota closed models don't even answer cybersecurity questions lol.
I think cursor will likely do a grok fine tune rather than a kimi one for the next composer.
they noted in their blog post they didn't focus purely on coding for grok 4.5.
Most likely, since they were acquired. But for us outsiders it would be a cool thing, to see if the delta is the same between kimi2.x + cursor data -> kimi3 + cursor data.
So nowadays the hardware and hosting providers must be in an optimization race, whoever can make the model just a bit smaller or more efficient (to fit on fewer/less powerful cards) will have a huge advantage and can make a lot of money.
I am curios what's the most profitable thing to "plant" (agriculture analogy) on the land (cards) that you have have: web hosting, vps, llms, image/video generation, etc
> if "labs are subsidising tokens on API pricing"
> SemiAnalysis estimates that Anthropic's current blended gross margin has risen to the mid-60% range, with the API business gross margin exceeding 80%
Of course, people will insist "they are lying", "why should we believe them, it's well known they subsidize API pricing", ...
https://newsletter.semianalysis.com/p/anthropic-3q26-profit-...
https://finance.biggo.com/news/02d45650-b569-4d12-b44d-8d6d8...
Agreed. My (somewhat educated) guess is that top labs have healthy margins on API pricing. But this release will add another 3rd party / clear of conflict datapoint in this estimation.
even deepseek, with their current (dirt cheap) price, can earn enough profit to cover the cost (hardware investment?) in 10 months.
Will the model even be competitive in 10 months though? Seems like models that reach top 20 on OpenRouter see 50% of all token spend by day 80, and 80% by day 180.
That numner blends in training or no?
Anyone who thinks that the labs are not profitable on per token API pricing is delusional and hilariously wrong.
It all depends if you count the fixed cost of training or not. And the cost of the hardware.
Or even the basis of the cost of hardware. There are lease deals, capacity traded for equity, various programs by Nvidia, there's absolutely massive depreciation, etc.
Getting 404 on the OP's link. Does it mean it got banned or self-censored in the meantime?
Until a few minutes ago there was a countdown page. (The weights haven't been released yet.) The countdown should be over in 19min, not sure why we're suddenly getting a 404.
I feel like most hardware to run LLMs on is shaped wrong for individuals.
It's either having a model struggling along with like 5-10 tokens per second on unified memory, or data center cards with hundreds of GB of VRAM consuming more than a kW of power. It doesn't seem like there's prosumer GPUs with like 180W-250W TDP and 128 GB or 256 GB of VRAM (one can dream). Then bifurcation and even just two of those cards would be kinda useful (albeit NVLink or equivalent would need to be commonplace).
Obviously nobody is running Kimi K3 locally without an insanely beefy homelab and lots of money to burn, but running GLM 5.2 would be cool at like ~100 tokens per second for a single session and maybe ~60 tokens per second with N subagents.
How unfortunate.
I have found that the "mostly didn't lose anything" Q8 large models that I want to run are all too large to run on the "only $3995!" 128GB max RAM systems that some people are buying, and definitely won't fit with any usable amount of context. Things like Qwen 3.5 122B Q8 or deepseek v4 flash Q8, or Laguna S 2.1 Q8 need 170-190GB of RAM including full context, which fits on a 256GB RAM dual socket workstation or rackmount server (sans GPU).
Copy and paste below from my notes and reported memory consumption with latest llama-server, assuming use of "--no-mmap" to load the entire thing into RAM at the time that llama-server launches.
DeepSeek-V4-Flash-UD-Q4_K_XL via unsloth 145GB on disk GGUF 0.03.323.204 I common_params_fit_impl: projected to use 178175 MiB of host memory
DeepSeek-V4-Flash-UD-Q8_K_XL via unsloth 151GB on disk GGUF 0.02.215.885 I common_params_fit_impl: projected to use 184636 MiB of host memory
Laguna-S-2.1-UD-Q8_K_X via unsloth 120GB on disk 0.01.616.119 I common_params_fit_impl: projected to use 172860 MiB of host memory
Qwen3.5-122B-A10B-UD-Q8_K_XL via unsloth 160GB on disk GGUF 165GB RAM use on launch, fresh context 0.04.976.905 I common_params_fit_impl: projected to use 170038 MiB of host memory
DeepSeek-V4 should use only 5GB for context due to CSA and HCA, see figure here: https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro
But not every framework implements it properly yet.
Yeah, or close enough to 5GB for estimation purposes, for example a just spawned qwen 3.5 122B llama-server instance reports as:
0.07.015.888 I common_memory_breakdown_print: | - Host | 170038 = 162913 + 6740 + 384 |
The 6740 is the cache size.
There's an emerging practice of using Q4 quants and Q8 KV cache for local inference. At that point you can run both Qwen3.5-122B-A10B (my personal choice on Framework Desktop 128gb) and Laguna-S-2.1.
Now whether that's good enough for one's use-case remains to be determined. You can also get more out of those (local models and quantizations) if you further tweak the harness you use them with, but tbh this is where it gets too much work (at least for me and the time I have available).
> emerging practice of using Q4 quants and Q8 KV cache for local inference
That's not an emerging practice, it's a tested strategy that is these days only used as a last resort by those desperate to fit a model in memory. Some models do better than others, but generally the model quality suffers greatly under those conditions.
I have never seen anyone report "this produced really great results" from intentionally quantizing their context vs. leaving it at full precision which is the ordinary default.
Gemma's QAT is surprisingly good (although Gemma isn't that great to begin with).
LLM inference unfortunately also seems to be a task that's poorly formed for moderate consumer hardware,as a single user. For a single user use case, the load is bursty but requires the weights to be in memory already. So a multi user server that keeps the model weights in parts of its memory and then spends some more per user kv cache is wildly more efficient and the wildly expensive gpu cores aren't just sitting idle most of the time. Don't get me wrong, most desktop workloads are bursty, but the power needed to to them has gotten cheap enough that we can have way overkill for idle scenarios hardware just sitting on our desks.
Why does it have to be so bursty though? Just let it run multiple continuous-batched inferences overnight. This would work especially well in combination with SSD offload, and given any kind of sparse attention (common in more recent models) even swapping out the KV cache itself to disk might ultimately be a win. I wouldn't be surprised if something like that ultimately became feasible for single users running even K3 itself.
I find it difficult to always have one or more long-horizon tasks 'queued up' and ready to run... I find myself usually bottlenecked on design, review, or something similar that requires me being in the driver's seat. It's possible I could queue up a bunch of tasks, letting the LLM run off in multiple directions, but then I'd be less able to steer and course correct.
Just my experience though, I'm still figuring things out. Perhaps some subsets of tasks would be more ideal for these long-horizon workloads - exploration, multiple competing implementations, etc...
Use your favourite harness to help you find long-horizon tasks to have queued up. It's changed the structure of how my projects work a bit, and do you have to do some homework fast of how slow/fast your various providers or local inference are, but it's worth it. Start off with hobby projects so you get a feel for how it works.
I left something gargantuan running over the weekend (decompiling 1980s-era system software) and look forward to checking it out later today when I have a few free minutes.
In principle, a slower inference ought to be easier to steer and course-correct. You'd always be able to look at partial results, especially with a local model that doesn't hide its thinking.
it has to be so bursty for realtime usecases like chat, which is what most people are using it for today. of course, once (if) stuff like software dark factories start working out for the average person, then you'll be able to make full use of your hardware for workload where asynchonous execution is feasible and have it run several parallel tasks overnight, with an orchestrator managing the gpu(s) allocations.
Chat doesn't have to literally be realtime though, that's just the model most users have settled on. You could fire off your request, let it work unattended and check back on it later (perhaps after getting some notification from the chat frontend via RSS, Web Notifications API or similar that the full response is ready).
If your workload fits long batches throughout the night you could schedule them better, yeah. But I think very few have a usage pattern like that?
Given the hardware shortage in the world, I suspect renting ("sharing") via APIs will likely remain cheaper for the foreseeable future since each piece of hardware isn't sitting idle nearly as much.
I've been feeling for a while that as we keep increasing model size, we're going through the opposite of the PC revolution.
The "democratisation" talk from the frontier labs is especially egregious when they only release closed models (gpt-oss hardly counts) and are trying everything they can to make it harder to run open models.
RTX Spark does go up to 128gb No idea how it compares tho
Yeah, I've been experimenting with DiffusionGemma which sadly isn't as smart as Gemma itself, but holy hell is it FAST, and has image input as well, so doing things like "take a screenshot once per second + ask the model to categorize/model it WITH reasoning before" becomes realistic and doable.
I ended up implementing DiffusionGemma myself with Candle in Rust + CUDA, and it's quite literally the fastest model I've managed to run on my hardware.
The next mac Ultra will allow to run a big model locally with acceptable speed. But we need people to optimize it for that computer, and we’ll be more limited in models we can choose from
128GB is enough to run a large model, quantized, REAPed, with MoE and fast SSD for model weights
Not Kimi K3 large though
> like 180W-250W TDP > running GLM 5.2 would be cool at like ~100 tokens per second for a single session
Your power consumption estimates are off for this generation of GPUs. A 27B dense model gets 50-80 tps on an RTX 6000 using 600 watts.
An AMD R9700 gets 20-50 TPS at ~300 watts on 27B. 100 TPS for the 35B MOE model. And there might be some more optimizations to that as AMD software support gets better with ROCm's latest versions.
Which quants? I get these speed (only 20-30 TPS) at Q4_K_M with MTP for 27B on my framework desktop. I draw sub 130W for the whole machine.
I agree but worth noting that it's never gonna be very practical to run LLMs like this at home. Unless we have some sort of design breakthrough, the only "sensible" way to run them is at high batch levels on shared HW.
Like, yeah if I could spend a few grand on such a GPU I probably would coz I'm a rich nerd, but I'd acknowledge it as an extremely inefficient luxury, kinda like a sports car.
So I think you could say the real misfortune is that we don't really have the technology (be it computer tech or political/social tech) to do that shared-HW thing in way we can truly trust.
"Never" is a long time. Just think about how much ram we had 10 or 20 years ago. 1.5TB isn't a lot really.
The typical ram has surprisingly not increased very much in 10 years.
> April 2016, 8 GB was standard across the 13-inch MacBook Air range
... Now it's 16.
Rich nerds will have quite a bit more. But I suspect the standard of model rich nerds want to use will have gone up somewhat too.
We could make LLM inference 100x cheaper to run at home efficiently, but that solution might need to be updated every 1-2 years, whereas current GPU are useful for various others tasks and last longer
‘Never’ is a big word in the computing world. 10 years from now a model this size will probably run on a high-end phone.
Of course, by then we’ll want to run something commensurately larger.
You're right, and it's interesting to consider why. It's probably a combination of a few factors:
1) Local LLMs are a relatively new phenomenon and hardware takes years. Apple probably lucked into their unified memory architecture being suitable (in terms of memory size and bandwidth) for local LLMs, but it's only with the newest generations we're hearing about LLMs even being a consideration in their design process.
2) NVidia seem to be deliberately blocking consumers from taking this path - as evidenced by the removal of NVLink from the 30x0 series onwards - probably to protect their data center cards from internal competition?
3) Perhaps there's just not the market for it? It's feasible that the number of nerds interested local LLMs is very small in numbers, sales, and profit potential compared to gamers on the one side, and data centers on the other. (This would explain why AMD and Intel aren't trying to out-innovate NVidia in this area, despite it being an obvious opportunity.)
There will be a huge market for local inference once it's cheap and widely available.
Try to imagine output token speeds of 15,000 tok/s and a time-to-first-token of 200ms. (This has already been done for Llama 8B.)
Now imagine gargantuan context windows (2M, 4M, or even bigger); keep in mind the 1M context windows were science fiction a few years ago... now imagine having this on a local model on something like a phone or portable device that can be gathering data about things you're doing and constantly run inference for things useful to you. An obvious example of this would be a chatbot you can talk to that responds like a normal human conversation and doesn't have delays, but that's just scratching the surface.
The individual-shaped-hardware problem gets even sharper at the phone end. Shipping a 3B model on-device, the usable RAM budget after the OS and everything else is more like 2-4GBtotal, not per-model so it's not 'can I afford more VRAM', it's 'can I fit a language model and an STT model and embeddings without the OS killing my process'. Feels like phones are the most hardware-constrained 'individual' tier and get the least airtime in these kind of discussion. Is that because the models that fit are still too limited to be interesting, or something else?
Recently spent a few hours messing with bonsai 27B and ternary bonsai 27B at total weights + context fitting in slightly under 6GB RAM, and it's just dumb as hell. It writes what seems like grammatically correct content but it's extremely limited.
It will also happily hallucinate new names and content to fill in gaps in its knowledge, and present the hallucations in what looks like a correctly formatted sentence, so it could fool a person who doesn't know the subject matter. Like, I asked it for a description of Seattle and it hallucinated a name and description of a nonexistant tallest building in the city and suggested the view from its observation deck .
https://prismml.com/news/bonsai-27b
https://huggingface.co/prism-ml/Ternary-Bonsai-27B-gguf
In no way was I surprised, it's asking a lot of under 6GB RAM usage. But I think for 99% of people they will get better results doing something over the network where the weights and inference engine are not on the device.
We already know that competition brought GLM 5.2 prices down roughly 45% since its release on June 16th (1.5 months ago), and the price downward slope is probably still going (I've been checking regularly and new providers keep fighting on price, I don't think prices have settled yet). For reference : https://openrouter.ai/z-ai/glm-5.2#providers
I saw arguments like "Providers cannot price less than their costs" in other comments. In economics, it's generally admitted that they shouldn't price less than their marginal costs, i.e. in their case roughly the cost of electricity, since a lot of these datacenters are not at capacity in terms of graphics cards usage (speculation since it's very easy to rent a GC for a couple hours on some providers). My guess is that someone will be selling tokens at less than electricity + depreciation of GCs soon, since there's a lot of competition and "smaller" data centers have overcapacity? This is speculation, correct me if I'm wrong
> My guess is that someone will be selling tokens at less than electricity + depreciation of GCs soon, since there's a lot of competition and "smaller" data centers have overcapacity? This is speculation, correct me if I'm wrong
My guess is they are selling you the tokens, then selling your tokens (data) onto someone else.
I see these conspiratorial arguments all the time and I think people massively overestimate the value of the average users tokens.
The problems with frontier models (design taste, ability to solve novel/difficult problems, etc) cannot be solved by throwing more slop from the average user at it.
Actually, most of the main deficiencies in current models stem from the fact that their data sets aren’t curated and specialized enough.
I don't think the goal of this data is necessarily model improvement.
I think it's marketing, advertising, and product refinement.
Ex: all the things Google wants your search data for.
It's somewhat silly to think the value of that data has changed much. Advertisers want to know what's popular and getting clicks and attention. Competitors want to know what features are getting used in their markets.
In the simplest case, think of this data as improving the harness, not the model.
The prompts contain sensitive personal data.
That would be valuable to advertizers for example.
Press x to doubt on the 45% number. The cheaper providers on open router are fp4 vs fp8 for official zai. There are some cheap fp8 ones (like novita) but the ui makes it seem like it's a temporary promotion, with their normal prices being almost equal to official zai (idk much about open router so not really sure what's going on with these discounts)
If you click the provider it shows the precision, 45% off at AkashML shows FP8. The drawback is the small context window, at 96k.
Then there's 43% off at StreamLake with FP8 precision and 1M context window.
There is nothing to doubt, the cheapest price on openrouter is ~45% lower than when GLM5.2 was released.
In my opinion, next step is to cut down on reasoning tokens while maintaining intelligence. The Chain of Thought and looping can still be an issue with these Chinese models. They in fact said K3 would improve in the area but it's still an issue that unfortunately harms the token cost wins a bit. OpenAI has been really impressive here, on the opposite end of this.
There is a really interesting startup in Prague that is doing just that. They fine-tuned Qwen 3.6 27b to have 46% fewer reasoning tokens while maintaining most of the performance characteristics. I'm interested to see if they continue down this path of optimizing reasoning for other models.
https://bottlecapai.com/post/thinkingcap-qwen3-6-27b/
Yeah, they thing forever and doubt everything "wait but" for 200k tokens for almost any question.
On the flip side, I really like being able to inspect its reasoning chain thoroughly, as opposed to the "black box" that Anthropic models are now.
Right I was going to say, no way of knowing whether these issues are unique to Chinese models.
Depends on whether the models report the correct amount of tokens.
5.5 Sol repors 10x fewer reasoning tokens than Kimi k3. If it is correct, than it unlikely has those doubt issues.
At the same time, I feel like their reporting is incorect and we are now paying per "intelligence", not actual tokens. We can't verify it anyway..
Given the frontier-level capabilities of Kimi K3, I'm wondering if it's possible to extract the core capabilities (fundamental reasoning and tool calling) of the model into a smaller one that consumer devices could run? Not sure exactly how, but either by heavy distillation or some other surgical method since Kimi has a Mixture of Experts architecture.
I think it's very valuable to have a smaller model that doesn't have any domain knowledge or facts built into its weights, but given the right context, could accurately reason about what to do and use the right tools.
I'm aware of colibri [1], but so far I've only seen extremely slow performance.
[1] https://github.com/JustVugg/colibri
"I'd like a car that goes 300mph and gets 100mpg while doing it. I'm aware of a car that gets 100mpg but it is extremely slow."
You are describing fundamental tradeoffs. Getting more performance relative to model size and training token amount is what all of the labs are solving.
Labs are focusing on creating models, small or large, that perform well on various benchmarks, including general knowledge, domain-specific expertise, and agentic capabilities.
Asking for such a model while wanting to be small and fast would align with what you're describing, which I believe is different from what I'm pointing to.
The model I'm describing sacrifices domain knowledge and expertise for agentic reasoning and tool-calling capabilities at a reasonable speed.
Think of Cactus Compute's Needle [1].
[1] https://cactuscompute.com/blog/needle
Why not generate an artificial dataset using commercial APIs and then finetune a small model on this data?
I’ve had success adapting even a 7B model for single-domain tasks that way, including reasoning and tool calling.
You can use an open model. The point is just to outsource the inference, so you don’t have to deal with running the larger model yourself.
This is probably one of the ways to achieve this. I see a plethora of such fine-tunes on HuggingFace [1], but they're either not much different than the base model or they're outright benchmaxxing.
[1] https://huggingface.co/models
There’s another way besides distillation that’s way cheaper: You can have the big model build prescriptive skills that the small model follows.
Take the “train” portion of tasks on some benchmark, have K3 complete it, and then output detailed descriptions of tools used and why, then run the validation tasks with some small model that has access to the skills.
Yes. Using a harness with a strong model to create lots of utilities and tools for yourself is effectively the same thing.
Isn’t that distillation ?
No. Distillation trains on a teach model's logits or output tokens.
K3 releasing as open source right at the same time that people are criticizing Opus for questionable performance (there's even a thread on HN about Opus 5's problems)... this is such a flex.
> (there's even a thread on HN about Opus 5's problems
is opus 5 a flop like 4.8 ?
where is the thread btw curios
Is there any (near future) technology that would permit burning this terrabyte into some kind of ROM chip?
Yes, from 6 days ago: https://news.ycombinator.com/item?id=48986351
A question of course would be "is 15,000 tok/s Gemini better than 100 tok/s Opus 5"?
Depends what you do. We have certain tasks we spend money on where Gemini 4.6 definitely is better than Opus 5.
It's better for the billions of free users that Google serves.
That looks promising! As models become a commodity, this may turn out to be the real AI gold rush.
Given how fast models are improving, burning the weights into actual ROM is prohibitively expensive if you need (or want) to replace the chip every couple of months.
The alternative is on-TPU flash for storing the weights.
Yes. We're still a ways off from this being ubiquitous, but I am convinced it's coming.
The path to ubiquitous AI (17k tokens/sec) https://news.ycombinator.com/item?id=47086181
You can still try it at https://chatjimmy.ai/, but it's running the rather outdated Llama 3.1 8B
They release it before I could even get a kimi subscription because of waitlist? lol hard to believe that I might get a kimi subscription from a third party
I just checked and I have it available in opencode go! I just tested it with one message and confirm it works.
It's not practical to use it though. It counts almost 8x more towards your quota than glm 5.2!
https://opencode.ai/docs/go/#usage-limits
Less than 2 hours left to the Kimi moment. It’s been more than 2 years since the DeepSeek moment that shook the world.
Coreweaves gonna boom
I heard this is the talk in town these days. Why can't Meta keep up? With >10000000x more resources you'd think that they'd be able to introduce equally performant if not better open weight models
The SemiAnalysis piece on this is long but very much worth reading:
> The company appears burdened by far too many disparate groups that are over-optimizing for certain metrics as opposed to delivering usable technology for the company as a whole.
> And because Meta has a reputation for throwing money at problems and executing at high speed, these U-turns end up becoming more costly versus other companies that take a more disciplined or conservative approach. Suppliers also lose faith when given design wins are later cancelled. This has lead to less supply chain prioritization on new designs. Some suppliers favor focusing on Amazon or Google designs due to Meta’s frequent reshuffling.
> Few inside Meta’s chip division have a full understanding of why the company bought Rivos in the first place, and those who championed the deal internally have since gone quiet.
etc etc
It goes into a lot of depth.
https://newsletter.semianalysis.com/p/metas-infrastructure-t...
he seems to be bullish on meta though and considers muse on slope than an intercept
https://newsletter.semianalysis.com/p/the-future-of-meta-sup...
The cynic in me says maybe they would be further along if they hadn't spent $80 billion on trying to build the "Metaverse" VR world. I've never met anyone who actually uses it and to the best of my knowledge it has very low mass market uptake.
https://finance.yahoo.com/sectors/technology/articles/mark-z...
Not exactly the best use of dollars and the labor hours of some of the best minds of our generation.
Particularly when you could vibe-code Metaverse VR for a lot less than $80bn.
Turns out sitting quietly in a room and doing math is worth more than all the money and network in the world.
Because lack of talent and organizational disfunction matters a lot more than you think. The reason why OAI and Ant are always at the top is because of this and I’d say compute is third on the list.
I would argue that they actually don’t lack talent, they have an insane bench of really smart people. What they lack is any sort of direction and leadership. They are a ship lost in the ocean and up until now have been lucky to find a few treasures along the their way.
> they have an insane bench of really smart people.
Filtered heavily into those who care only about money. Many don’t want to work there. Smart people have other choices.
because imagine starting work every week and finding that your dumb ass CEO pivoted the company again and is ruining other peoples lives, and he then reorgs the management again so you now have your 5th leader this year.
Morale and momentum are huge things in companies, Zuck has been murdering both of those in Meta since... well naming it Meta.
At this point in time what does meta get from releasing open weight models? Why devote the resources to it.
Why would there be any motivated person left in this place.
You can make the same argument for closed models. Why spend hundreds of billions training larger and larger models when you can just use Chinese models? Spend that money somewhere else further up the stack where there’s more value. Let China do the training since they’re so efficient at it.
Because without bigger players, Chinese models don't get anywhere. Playing catch-up is a radically different game.
as a big tech company you have the resources to make many bets and do a lot of things at the same time. it's good to have some specialists with knowledge of model training "just in case".
Even Meta, with all their resources, makes all their hardware in China. Manufacturing anywhere else is just burning cash.
Did someone run censorship and political bias tests on this ? Must be interesting.
As a completely one person, single sample anecdote, the 'heretic' uncensored Q8 GGUF variants several people have published of Qwen 3.5-122, 3.6-27B and 3.6-35B-A3B will very happily discuss just about any controversial topic that the CCP hates. Including lots of things that would get you thrown into prison if you published them in Mandarin on the domestic Chinese internet.
https://github.com/p-e-w/heretic
As a side note on this, if you see the reference in the screenshot in the link above to the harmful behaviors prompt set, these are all in English:
https://huggingface.co/datasets/mlabonne/harmful_behaviors
You could likely further de-censor a model by having a set of 'test' prompts in native Mandarin, Cantonese or really just about any other language. I don't speak any Chinese languages so I don't know if the published 'heretic' GGUF files some people have been throwing around will cooperate, or refuse, if you ask it in Mandarin for how to build a meth lab or precursors for semtex.
That's a lot of words to say "No, no have has seemingly done that yet with K3".
Indeed not, but I was saying there's more than ample precedent which is tested/working and actually doesn't refuse anything. Go to the Huggingface 'models' search interface and type in "heretic". Or uncensored. One example would be: https://huggingface.co/HauhauCS/Qwen3.5-122B-A10B-Uncensored...
Right, but aren't we jumping into trying to figure out solutions before someone actually checked if any solutions are needed in the first place?
Yeah, I think we will know more within a couple of days, once people actually download/run/test it. I'm sure the people adjacent to the 'heretic' developers will give it a test as soon as they get their hands on it. All very theoretical right now.
> I think we will know more within a couple of days
It's a 3T parameters model, with a weight format (MXFP4) still not completely integrated into the ecosystem, which only a few has the hardware to even do inference with, much less fine-tuning or more post-training. But sure, do sit and wait a few days :)
Yet the comment was valuable nonetheless.
Outside of asking it to talk about Tiananmen Square, are there any standard tests for "bias"? And if so, who created them and what are their biases?
There's is always one in each and every AI forum: "But, but have you asked it about tiananmen"?
It would, indeed, be interesting to compare, given what we know about Anthropic’s censorship and political bias in their closed and more expensive models.
https://x.com/dhh/status/2081435006770249831 (from the creator of Ruby on Rails).
In his specific case Kimi did the task it was asked to do (translation of the article DHH wrote), which Claude refused to.
we need open models because they let me dehumanize roma ppl. great take by dhh.
The article did nothing of the sort: https://x.com/dhh/status/2081435971678261344
However, even if it did, it is absurd for an AI to decide what it will and won't translate.
It's 404 now
This is historic. For the first time, an open-weights LLM is right at the top.
We won't be able to run this ourselves, but many providers can.
> For the first time, an open-weights LLM is right at the top.
Hmm, not quite true, I think that honor, for better or worse, goes to OpenAI. When they released GPT2 (or GPT1 for that matter) is was quite literally the SOTA in the ecosystem when it was released.
You are correct. I miss the time when OpenAI was open.
When OpenAI was actually still a proponent of open AI...
thank you embedding-shape
That would be 7/27.
China uses YYYY/MM/DD, which is logical.
What China (and Japan) uses is YYYY年MM月DD日, which IMO is the superior date format since it's self-explanatory - the sections are spelled out right there!
the only logical format.
signed: a hungarian :)
For me, the only format that doesn't make sense is the MM/DD/YYYY, together with its rarely seen worse sibling, MM/DD/YY (07/27/26).
Let's just go with YM/DY/MD (27/26/07)
But only for Americans, and make them weirdly dogmatic about it.
Agreed. It's objectively mixing the order (medium/small/large)
You can sort dates and they are in order with this format.
lpszReleaseDate ;-)
LoL
You mean a Big Endian.
20270727 if we're improving dates :)
Jumbling together year, month and day and having to separate them by counting digits is not an improvement for human readability. (And you have the year wrong.) Any of “27 July 2026”, “July 27, 2026” or “2026-07-27” would be superior.
It's good because it's an actual standard instead of 7/27 which is backwards for half the world. Besides, I'm actually from the future and they delayed the model a year, learning from Anthropic that good models are actually about to bring the end of society.
ISO 8601 ftw
270727 if we want to save tokens :)
Y2K100 bug for you, then ;)
Looks like another LeetCode problem about checking for anagrams.
or 27/7 for the rest of the world
Most life forms don’t use any calendar actually.
Don't even have to go that far, outside of white-collar jobs and some groups weirdly obsessed with scheduling, most people don't use calendars at all, but their manager/boss/significant-other does that for them :)
No, 27-7 for the rest of the world.
The separator is often the only way to distinguish American notation from ISO, so please use a dash for dd-mm-yy and a forward slash for mm/dd/yy
Have never seen 27-7, as someone in rest-of-world
In my time it was “27/VII 1986” in Russia.
This is so confidently wrong it's funny. In Australia dd/mm/yy is the default.
I've never seen dd-mm-yy. It's usually dd.mm.yy, dd.mm.yyyy, or yyyy-mm-dd, with some dd/mm/yy sprinkled in for general confusion.
Or 27.7. in some other places.
Nope,
There are quite some countries around the world using d/m/y
https://en.wikipedia.org/wiki/List_of_date_formats_by_countr...
Algeria, Belgium, Brazil, Chile...
27–7 would be 20.
Not really. Spain's traditional format is dd/mm/yyyy with slashes. This applies for a good chunk of Europe. Germany/Austria uses dot, I think nordic countries embraced the dash. While you might see more adoption in offical/digital contexts, I just double checked a few popular spanish websites, all slashes.
You mean 27.7
Or 11/Shrawan 2083 if you're in Nepal
This has to be one of the craziest uploads on the internet up until now.
Raw fucking intelligence at your disposal, free to download.
If you'd describe what's happening now to someone from five years ago they'd think you're hallucinating or mad.
Can't wait to run this at 0.02 tokens/sec on my CPU so I can get a response just in time for next month.
Unfortunately I don't think modern consumer CPUs are physically capable of addressing enough RAM to even load the model into memory. We'd have to wait for some random person to make an extremely quantized version before we could reach those blazing speeds
I wonder how long it will take for this to get fully decensored and for bad, BAD things to happen
Are they going to release Kimi K3.1? I’m eager to test it. According to rumors on X, it could outperform Fable. Could it be the first Chinese open-weight model to become the leading frontier model?
is there a realistic way to distill 2 consumer hardware friendly models with max ~200B and ~20B? Qwen did it, but would it be possible for 3rd parties (unsloth etc)?
Yeah, why not. Toughest part is running the hardware so you can create the traces for downstream training, but once over that hump, nothing would stop you from doing that no.
Hoping no issues on Huggingface due to download rush.
For huge models like these, the only reasonable way to host them is via torrents. I don't understand why hf doesn't offer this as an option.
Linux distributions got this right: Offer both HTTP and Torrents. Let the user decide.
I believe they don't do torrents because it gives them a lot more control.
Like they can takedown or update downloads and they can prevent someone from trivially bypassing the license agreements you need to accept for some models.
> Let the user decide.
Perhaps that's exactly what they're trying to avoid, giving the user any form of control and having them depend on HF.
It’s going to get very little downloads just by virtue of size. Very few shops will be able to directly use it
This looks really promising. Excited to see where this goes. Looking forward to trying it out!
Good enough is often the right call
I don’t need the fastest car to get to where I’m going. I need a car that gets me to where I’m going at the speed I’m comfortable driving at.
Why is there a countdown?
You're not having a party?
I think it's shameful that Moonshot isn't providing us with party kits like Microsoft did with the Windows 7 Launch Party kit. How am I supposed to properly celebrate this without fun Kimi-themed quizzes for my guests?
At least they don’t make you stay till the end of the presentations to give you the software you actually came for
It saves you from having to perform date-time calculations.
It's a release party
maybe a quantized version on a GB300 would work? unsloth hopefully working on it.
The VRAM, power, networking, and operational requirements put it beyond the reach of many enterprises.
The weightings should be released on July 27.
how feasible its will be to run on modal or deepinfra? anyone here tried and tested such large models running?
There’s going to be a lot of competition around this model. Let’s see how low AI providers are willing to push prices.
I think the results might be underwhelming - AI providers need to turn a profit and can't subsidize, and they're working off of the commodity hardware everyone does.
I wouldn't be surprised if they started offering potentiall bad quantizations with much reduced capability at lower prices (without telling the users, of course)
I would be surprised, considering that OpenRouter requires disclosing the quantization and shows automatic benchmarks to compare between providers for the same model.
The latter is a joke
I saw that it runs GPQA Diamond and TAU-Bench Airline and shows the results over a 32 day rolling average.
Other than that they track Tool call error rate and Structured output error rate.
I only discovered this today, and it seems like a good idea. What are the problems in practice?
As long as they are transparent about what quant they serve the model and any other optimization they do that also affects performance of inferred tokens.
...does Hugging Face have enough bandwidth to let people download en masse however much file size a 2 trillion parameters model is?
I have a feeling the number of people downloading 2t parameter models is significantly smaller than the 1-100b models.
There's no going back on this. This is putting a very capable intelligence in the hands of the masses. Private companies in the US are aching for Trump's protectionism but it'll do nothing. The hardware needed to run this is ofc prohibitive, but actually putting it out there feels like a 'RSA source code on t-shirt' moment for humanity.
I don't know if the masses can quite afford the 500k in GPUs you need to run this
No, luckily private companies in the US are aching for this to NOT be banned. NVIDIA, Microsoft, etc. just released that letter. We’re saved from the trillionaire companies (OpenAI, Anthropic) by the other trillionaire companies acting in self-interest (hosting and hardware).
It's thankful that openAI or anthropic haven't IPOed yet. Or terrible for some
Strange communists, giving away such an expensive model to the public.
On the other note, can't wait to see 1bit quantisation soon and how it performs in benchmarks, if it performs really well in benchmarks, would be very good news for GPU hosting providers, to offer "Opus 4.5 level model at the cost of Haiku 4.5"
Now I hope that nvidia will host it for free :-)
if it's anything like the speeds Nvidia Nim puts Deepseek at, it'll probably be 10 tok/s or lower and timeout frequently
So... Now we give huggingface the hug of death - right? ;)
This is (actually) AGI, that truly benefits all of humanity with zero gatekeeping.
Now the US government has 5 hours left to (attempt to) stop the release. (and save Anthropic)
Let competition run its course and the market (not government) determine the winners and losers.
Will it work on 4GB of VRAM? /s
Wait, so I can download it and run it locally now?? Wow... But it probably won't work on my computer, right?
The short answer is no, it won't work on your home computer. In it's current form it needs something like 594 GB of memory, far outside what you can reasonably run on normal consumer hardware in 2026.
If you have really high end hardware, you might be able to squeeze a heavily quantized version of Kimi-K3 onto your rig, but it will be too slow or too lobotomized to be useful.
This does put a near state-of-the-art open weights model within reach of what a small or medium business could afford if there's a case for local inference. It's probably not as good as Claude Fable or ChatGPT Sol. But if you're an organization that has a genuine need to run inference locally, this is a real possibility.
Is this for your homelab? Not in any practical sense.
Is this a possibility for organizations that can justify $1M or so on hardware for a near SOTA model they have full control over? Yeah, absolutely.
The full K3 model will probably be way more than 594GB, that's more of a plausible range for Kimi 2.x. You'll probably be able to test run this model at full or near-full precision using SSD offload, but only at very slow speeds - probably slow enough that you'll be forced to let inferences run overnight or even spanning multiple days. Mind you, that's still useful enough for many casual users, given that they're running a near-SOTA model!
Yes it will. Buy a 4TB nvme, allocate 3TB as swap, and run a gpu emulator on your cpu.
I can't imagine a GPU emulator would run better than straight CPU.
Right
FYI huggingface refers to the alien from the Aliens movies that we need to prevent from reaching earth at any cost because it means the end of civilization.
Just checking in because y'all sound good with that.
Those are facehuggers.
Doesn’t it refer to the emoji?
haha stupid xenomorphs with their acid blood and pointy bits - all they had to do is make the beasts write code and do our homework :)
When Gen Z is in charge of security protocols, lol.
Funny retcon, but come on… TIL huggingface started as a chatbot for teens who didn’t get enuf hugs.