MiMo v2.6
mimo.xiaomi.com
mimo.xiaomi.com
The realtime dashboard they shared during training (https://mimo.xiaomi.com/rl/) was an incredible learning and teaching tool for me, and they’ve been unusually comprehensive in sharing details about their methodology (check out that tech report - it's got lots of clever behind the scene tricks like Google or Deepseek writeups) and benchmark scores (even the stuff they didn’t do well on).
If you’re releasing an open model going forward, please consider offering the community more of this transparency!
Really the only thing missing was dataset descriptions, the dashboard only had random IDs like "dataset-zrso". I guess it's their lawyers fault.
It'd be great to see a description of even just a subset of training datasets. It feels very much under-reported how much expense is worth investing in preparing and selecting training datasets versus just using masses of random quality unprepared training data. This dashboard appears to be good though in showing the limits quickly reached when throwing parameters and compute at the problem.
For example, if they were to train on Wikipedia dumps, do they consider every article to be the same quality across each language, or have they done more work beyond Wikipedia's own article quality ratings to make training decisions such as "Ignore cebwiki it's machine-generated spam" and "Treat dewiki articles with coordinates within Germany as being higher quality (weight it higher) than their equivalent enwiki articles".
And let's say one of the datasets is all the source code of packages in the Gentoo package repository. Not every software package is a good example of how to write code. You perhaps wouldn't want to train your LLM on 1990s era PHP web application source code as an example of how to write code in 2026. Instead, you'd possibly want to use such PHP web application source code as a negative training example of what _not_ to write. But when training an LLM to detect software bugs, maybe outdated PHP source code is good for training.
Similarly for translation, perhaps UN treaty documents translated into 4+ languages are good translation examples because of high accuracy needed, professional translators being used, and bigger budgets. However this training data would perhaps be a negative training example towards translating chat messages, movie subtitles, etc because it doesn't use everyday slang and could result in output of nonsense such as "Pending Your Excellency's response, please accept, Your Excellency, my sincere greetings." for a prompt asking to write a birthday card for a child.
Preparing training data and deciding how to best use it for training I assume would be the largest expense (cost of labour -- mostly expert labour too) and also the greatest opportunity in the future for LLMs to improve. It seems to me somewhat irrelevant if the dashboard indicates a compute expense of $1m or $5m if good training datasets (prepared by experts in their fields) cost $10m/y to maintain. For example, hiring expert software developers to tag 1000's of open source software packages according to their quality, on different metrics, such as human readability, performance optimisation with choice of algorithms, reasonable trade-off between coherence and coupling in the software architecture, currency with state of the art programming trends/preferred dependencies/operating system APIs, etc. And keeping that metadata continually updated rather than a rapidly obsolete once off tagging project completed in 2005.
Here is one really neat bit:
A cutting edge training idea (for agents, it's been used elsewhere for ages) is on-policy RL, basically, it's not enough to say "here is an end to end agentic sequence (including tool calls etc.) that is perfect" you want to say "here is a sequence you might actually have generated that turns out to be correct".
Basically, it's more training efficient to improve models with small tweaks to do more of the right thing they are already doing sometimes than from some perfect oracular "this is the way" answer.
(if you've ever tried to teach humans new skills, you’ve probably noticed this too!)
When you do that, you care about how far the model you are updating (improving) has deviated from the one being used to generate rollouts (agentic rollouts for hard problems can take hours with lots of tool calls, so you can't keep redeploying every slight improvement).
Lo and behold, the dashboard literally has:
partial/avg_staleness (likely the measure of how many micro iterations the "generate answers" model is behind the "improving based on the occasional right answer" model)
train_infer_diff/new_infer/kl (a more direct KL divergence based way of measuring how differently the two models generate tokens)
How cool is that?!
And don't get me started on the clever ideas hiding behind dynsam/avg@n ...
whey they all singing the same tune. it make me question what is their real motives.
they are afraid of Chinese good enough LLM model killing their margin. we already have story about US companies switch some task to use cheaper Chinese model hosted on Neoclouds.
Both Anthropic and OpenAI leaders have repeatedly made this exact argument that it's impossible for open models to rigorously enforce the same kind of safety framework as proprietary cloud-served models. It's implicitly part of any regulatory framework they advocate or else it wouldn't be "fair" to American companies since Chinese models would "cheat" (provide weights).
This is already kind of the case: the big enterprises don't really want to touch the latest Chinese models. It's a real pain, personally, I want to use them at work!
2. Enterprise trends are towards open weights, several routers and vendors now have more than half the volume going towards open weights
His first tweet ever, from this last July
https://images.nvidia.com/pdf/Open-Weights-and-American-AI-L...
I do imagine it'll change, but it hasn't yet.
Until profitable, reputable third parties host open models in the US with ZDR or they become plug-and-play for self-hosting at a modest cost, paying the US models is as much about data protection and liability as performance.
It’s similar with cars. It’s not that American cars are better than Chinese cars on any tangible measurement. But America already shipped most of its manufacturing overseas. Everyone who built those factories is retired. The US should probably hold on to some capacity to make cars, seeing as their entire infrastructure depends on them.
See eg https://en.wikipedia.org/wiki/Friedrich_Merz#Private_sector_... for the current chancellor. Many past chancellors were also lawyers, and many members of the Bundestag were and are lawyers.
I don't know whether having lawyers in power leads to industrial decline. My point is only that you can't use Germany as a counterexample.
Local LLM's are hit worse. Its about 6k for 5090 or 15k for an RTX 6000 and the Mac Ultra 256 is upwards of 12k.
Sure if you already have hardware, you can frankenbuild a system - but even the "affordable" dev stations of the DGX Sparks went from 3500 to 5k and upwards of 8k depending on vendor.
All the meanwhile, OpenAI pushed Luna 6 which is crazy cheap suggesting they have flash models to compete with Chinese models.
I just hope we see more open weights. Nvidia has NEMO but their license doesn't allow NEMO to be re-used on say, Apple or ROCm - Nvidia hardware only. Big ol MEH
They are not proposing to regulate only the strongest models. They are proposing to regulate all models. If they are already on top, regulation may stop them from proceeding further, but it also stops the cheaper alternatives from catching up.
If they feel they have reached the asymptote of the curve, then regulation doesn't affect them, it affects those who have yet to reach the asymptote.
My guess is that Anthropic and OpenAI will push for "safety" regulations which require byzantine testing that, shocker, Anthropic and OpenAI can pass but the chinese models cannot. The route they'll take is import bans and potentially even general bans on products producing or using "unsafe" models.
They'll further likely try and push AI "safety" treaties from the US to other nations to further lock in their lead.
That's why, IMO, we've been seeing so many "OMG, AI will destroy the world and these AI researchers are so scared" articles.
What's different today is that now all the big LLM firms want to slow down AI development. When men like Musk and Altman (both known for habitually shooting their mouths off and saying whatever they need to whoever needs to hear it regardless of truth) suddenly agree with Amodei, that's when things start to smell off.
not all, just a few American ones (~PayPal Mafia + Google), there are other big American LLM developers (notables include Nvidia, Meta, and Palantir) that do not agree
They invested billions into training the models but there is no competitive advantage, we see that within a couple of months everyone catches up. There is no way to profitability unless they get some policies to shields them against competitors that can't comply with the regulatory requirements.
That is also why there are things like Claude, Codex and Cursor. They are trying hard to build a customer relationship with a higher switching cost that hopefully sticks.
But the problem is that the AI buildout has become a large percentage of GDP. So obviously the government wants to keep it going because these companies are pumping enormous amounts of money into the economy.
They are pumping enormous amounts of money into each other. Hardly any of that is making its way to people, it's all going to highly automated construction and to energy use.
Seriously, how many jobs did the $1t in venture capital fund?
If everyone goes around mowing lawns for each other, the economy is richer in lawn mowing at the expense of all the other things that would have been funded had everyone mowed their own lawns and purchased different services instead.
Person B owns shares in Person A.
> eru
Tolkien fan?
Tolkien is great, yes.
Value was created!
How do we know that? How automated is the construction really?
In any case, the Fed and other central banks can print as much money as they want in order to hit any aggregate spending or inflation target they have for the economy.
When eg I snap a picture of my dog, that's not derived from anything. But I still get intellectual property rights for the photograph.
Aren’t they alleging these other companies directly entered into a contract and violated the terms, and in cases where question, answer pairs were obtained without such agreement, it was accomplished by outright wire fraud or theft?
Besides, the latter actually published and open-sourced its RL stack to make it reproducible, which would in fact make it more trustworthy than the models you are speculating were distilled.
But even with all the open models the US is just insanely ahead in AI buildout and capital allocation (as usual).
Is it a bubble? Is it like the race for the-first-to-the-nuclear bomb? Both?
It won't be a great moat, they'll probably try and get trade treaties setup to try and expand the moat. But ultimately it won't slow down chinese model development, just limit who can legally use them.
From politicians like Bernie Sanders, we've had proposals like 20 year imprisonment for anyone researching "ASI".
So how could Dario show the investors very nice profit charts representing profit = revenue excluding training costs if it needs to pay a lot of training every x months?
They want to sell the same model for longer(a kind of software subscription where the cost of running /inference is cheap) but the Chinese don’t let them do it. That’s the gist of it. You can see already how they nerf the models just a week or so after release and try all kind of tricks to deliver you shitty performance for the same money. I think it’s part of the same issue of costs and enshitification plan.
In the meantime let’s hope they don’t get to ban the Chinese models(I think they won’t), local AI hardware will get cheaper and the whole AI doom saga will slowly fade to the point that Anthropic becomes a kind of IBM stuff with proprietary data, enterprise certified alignment and enterprise contacts. Think of Accenture junk.
For frontier labs the current compute seems to be driving model progress (in training) at least to some degree, even without true RSI, and this seems like it'll continue to keep any chinese model from drawing even with the frontier labs, at least for the foreseeable future.
Inevitably the chinese government will drive more funding in chip fab technology and the money will come around to build chinese data centers, but who knows how far off that is. A few different things in the tech tree need to fall into place. It doesn't seem like it'll be next year.
If there is no upper bound how how these things scale with compute, and if China does really begin to catch up to Nvidia (and they're probably not going to feel encumbered by US patents for domestic AI hardware, given how important AI seems to be to the Chinese government), there will come a day when China leapfrogs the US on AI.
For instance a Chinese EUV machine seems like it's very far away. Even if they have (steal/borrow) the necessary IP.
Chinese models are increasingly closer to the frontier, while being able to run on much cheaper hardware than what US frontier models run on.
On top of that, both Anthropic and OpenAI showed that they can't really be trusted on data security.
Even if US companies can be forced to not use Chinese models, the rest of the world is going to see the risks and the availability of good enough open weight models for their purposes and be more likely to lean in favor of self-hosted Chinese models or local inference clouds.
it is childish to believe Chinese are going to give up profits to provide such open weight models forever. the whole idea of being "open" is not compatible with the Chinese culture.
Chinese companies will continue to provide open weight models as long as it is profitable to do so. Chinese companies are on the more open end in many other industries despite the lack of meaningful foreign competition (for one, 3d printing) so there's plenty of reason to be optimistic as far as I'm concerned.
As normal consumers with common sense, we should never naively assume others care for the world out of the goodness of their hearts. Maybe they do, but we should never rely on that.
We can only get good, affordable deals when there is enough competition on the other side.
Xiaomi MiMo is led by Luo Fuli, a former Alibaba & DeepSeek employee. Perhaps it is due to Luo just how similar Xiaomi's tech & GTM approach is to DeepSeek's.
- How Luo Fuli Keeps an Earthy Touch as she Soars Through the AI World, https://newsen.pku.edu.cn/news_events/news/people/15385.html (https://archive.vn/I8Pmu).
- Luo Fuli, the 30-year-old ‘AI genius girl’ behind DeepSeek’s success?, https://e.vnexpress.net/news/tech/personalities/who-is-luo-f... (https://archive.vn/sb3B6).
I wonder if this waters down the “distillation attack” claims by Anthropic. They have their own RL environments! I guess the caveat is that the RL datasets are still opaque, nothing is really proved.
My dream is to see this like a dashboard for a model trained across distributed machines, like Bitcoin mining, where minted coins are given to people whose machines were used for training. I don't know if they are worth it, but bragging rights alone, like a tag they can put on a website or social media, will be good enough for me.
A) How to prevent malicious injection of bad training data?
B) How to handle copyright violations, will participants be responsible and will they have to pay the creators?
C) Can this create an income stream for content creators and how to avoid abuse, eg feeding with AI content?
So many more, but let’s focus on these before we break things fast because we didn’t think about them.
At the moment we have Annas Archive being paid by frontier labs and rare/second hand books being destroyed in order to support the training regime. If instead we could just pay the publishers and they could distribute royalties to authors...
I'm sceptical though whether use of LLMs would encourage creation of data that doesn't already exist. For example, if you ask an LLM "What are the top 100 most prevalent flora endemic to ACME National Park", this data may not currently exist _at all_, and to collect, would require paying botanists to do an extensive field survey. If no one has done this work yet--why? Is it relevant to the scientific community, to making government decisions, etc, or just an obscure academic curiosity. There are certainly some journal articles on _other_ national parks describing some of their common endemic flora, but perhaps there was a reason for this. Such as a scientist funded by a one-off government program trying to determine how to preserve or even create habitat for a specific endangered species.
Consider for the prompt of: "What are the top 100 most prevalent flora endemic to ACME National Park"
An LLM may reply: "I couldn't find any journal article or other prior work that may answer this question. Typically such survey field work may cost $X to complete, require expert botanists, and take 6-12 months to complete. Let me know if you want further information on how to find and select a company to conduct such a botanical field survey."
Would this type of LLM response grow the industry of botanical field surveys, or do nothing, perhaps because anyone likely to fund botanical field surveys is already doing so regardless of whatever is happening with AI.
This can be a fun conversation and discussion.
'twas discussed a few days ago:
Xiaomi Mimo 2.6 live post-training dashboard - https://news.ycombinator.com/item?id=49732270 - Sept 2026 (155 comments)
no idea why I laughed so hard at this, feel so familiar
I think you mean mostly the US
No matter how much cash you throw you can't just materialize a 100 nuclear reactors to power the data centers.
For processes that are scalable, well known and industrial, profit motive actually does provide results.
The question is more of are there bottlenecks of skill and maybe secret sauce (eg something like ASML).
Also, unlike China, US companies are building data centers all over the world, which gives them higher distribution and ability to colocate with the energy production sources.
Lastly, energy production costs have been decreasing over the last couple of decades. If they will increase, the market will react, as it always does. Looking backwards does not predict the future in this case.
As these models get smarter they will no longer distribute it openly. Patel reporting this too.
There are real headwinds that I don’t think people have thought through.
Yeah, right.
Like Terry Tao recently said, there are nonlinear effects at play. Things are going to get chaotic and I do not have confidence (like the parent comment) of anyone "winning".
I agree in principle, but it could be more than 5 years, maybe 10. Who knows what things will look like then.
Looking at Nvidia revenues in the past few years, there's maybe $300 billion worth of GPUs currently deployed in the U.S. The B200 costs ~$40k, so we have 7.5 million B200-equivalents, which draw 1000W. Running these at full capacity requires 66 TWh a year, or ~1.5% of total current U.S. electricity consumption. Maybe a bit more to account for inefficiencies, cooling, and other components, but not more than ~2.5% total I would guess.
So it's not that much in reality, but will definitely grow fast.
Hyperscaler PUE (which includes cooling, power conversion, etc) is typically 1.10-1.15, so multiply 1.5% by 1.1.
The person you replied to already mentioned cooling and multiplied by 1.67 to cover ancillary uses.
I think your numbers are off.
For a start you are effectively calculating a GPU only number.
I think 100Twh would be the minimum level to think about "all-in". And even that is probably being generous.
Remember, afterall that Google have just bought half the capacity (4.1Twh) of a nuclear power plant in Finland, on top of 630 MW of wind and 94MW of battery.
This is to cater for three new sites at Kajaani, Muhos, and Vaala and expansion at Hamina. So basically 3.5 datacentres.
But Finland is quite a small place. The US has more sites and bigger sites, so the numbers probably grow exponentially very quickly.
> "Quite frankly, the biggest issue we are now having is not a compute glut, but it's power and it's sort of the ability to get the builds done fast enough close to power," he told the show's hosts. "So if you can't do that, you may actually have a bunch of chips sitting in inventory that I can't plug in. In fact, that is my problem today. It's not a supply issue of chips. It's actually the fact that I don't have warm shells to plug into."
> - Satya Nadella
https://redmondmag.com/blogs/generationai/2025/12/microsoft-...
So that’s how the Matrix got started …
Aluminium is often described as "congealed electricity". Ship bauxite to wherever power is cheap and stranded, turn it into metal, and ship the metal out. Here in NZ, Tiwai Point is the textbook case, with London-based Rio Tinto running a smelter on the other side of the world that exists mainly because Manapōuri hydro had nowhere else to go.
AI data centres can be just the same - even more so, since the plant's assets (its chips) are virtually perishables, so there is less concern about assets becoming stranded if the host goes rogue. All the US needs is friendly and stable allied countries with cheap power.
Datacenters enable anyone with a computer to use it.
Plus another bottleneck - China produces engineers, the US produces lawyers.
> For the US, ASEE and NCES put engineering bachelor's degrees at a stable 130,000–145,000 a year, rising above 200,000 if computer science is included. Because China counts CS as engineering, the fair comparison is roughly 1.35M against 230–250k US engineering plus CS graduates. That is about 5–6x in absolute terms. Adjusted for population (1.41B vs 335M), China produces about 950 per million people and the US about 700 (1.3–1.4x more).
Even Claude highlights that the numbers aren't apples to apples and in China the definition of "engineer" is more loose. I do believe that in relative numbers, more engineers graduate in China
On lawyers:
> In the US there were 1,322,649 active lawyers as of January 2024. For China, there were around 650,000 lawyers in 2022. Which in relative terms is 8x in the US than China.
Lets see how these numbers stack up
Annual PHDS
CS: US = 2300 China = 5500 ECE: US = 2400 China = 7500 Math: US = 2200 China = 4500
Now of the US PHDS what percentage of the PHD's are awarded to students from China?
CS: 32% ECE: 35% Math: 26%
There is a strong chance most of the researchers are pulled out of the US and Europe if WWIII really kicks off or even if there is just more global crisis and concern.
One other thing about the power needs. Within a few years, the power efficiency of AI chips is likely to improve by a factor of 20, 50 or more times by switching to true compute-in-memory architecture with new materials that have made rapid progress lately.
for wwiii it would likely imply a war in asia too, so it's not as if PRC will be a safe place for those researches to run away to.
So it is very unlikely that a neighbor will have the audacity to attack them, and even if this happens, i'd expect the war to be over rapidly. Just like if Canada or Mexico tried to attack the US.
It's almost as if collaboration is the foundation of scientific progress. Too bad US has lost the notes.
There's a very american winners and losers, black and white, narrative spreading in every single field from decades.
We Europeans used to be more nuanced, but are slowly falling for it too.
https://www.wabe.org/utilities-in-georgia-southeast-may-be-o...
https://cleanenergy.org/news/psc-approves-georgia-powers-dan...
https://www.smarterdigitalvirginia.com/post/the-big-bottlene...
just to name a few, there's a lot more I list out here: https://news.ycombinator.com/item?id=49700743
Pro [2]:, 1.02T total / 42B activated parameters
Are others seeing the same thing?
This is also true for Deepseek 4(.1) .
Maybe Terminal Bench 4.0 and ExploitGym are reasonable.
Terminal Bench 4.0
GPT 6 Astra 59.6
Claude Fable 5.1 55.1
Claude Opus 5 49.0
MiMo-V2.6-Pro 34.9
MiMo-V2.6-Flash 28.8
DeepSeek V4.1 Flash 26.8
MiMo-V2.5-Pro 1.5
GPT 6 Astra 42.4
Claude Fable 5.1 30.4
Claude Opus 5 22.1
MiMo-V2.6-Pro 17.8
MiMo-V2.6-Flash 6.0
MiMo-V2.5-Pro 0.1
DeepSeek V4.1 Flash 74.2
Claude Opus 5 74.0
GPT 6 Astra 74.0
MiMo-V2.6-Pro 71.9
Claude Fable 5 70.0
MiMo-V2.6-Flash 67.9
MiMo-V2.5-Pro 19.0Maybe I am being routed to more quantised versions or less capable models with system prompt to fake Astra or Fable.
TerminaBench, DeepSwe sites are out of date.
Pelicans for Pro: https://tools.simonwillison.net/markdown-svg-renderer?url=ht...
Jokes asides, @simonw any plan to include a 3D model version (make a 3D model of a Pelican riding a bicycle in Blender)?
Given that Astra seems to have improved a lot in 3D modeling capabilities (and that matches my experience) I'm actually quite interested to see if/when other models catch up and how they stack against it.
Or if anyone knows what would be a useful existing benchmark for that skill.
I've not committed to that as a benchmark yet because you really need a full coding agent configured to run Blender, and I like benchmarks I can run as a single prompt/response through the appropriate API.
I have been intending to get more of a coding agent benchmark going though, so maybe this should be part of it.
Qwen3.8-27b pelican was amazing on Mac.
In any case, pelicanmaxxing comments without any further proof or evidence probably fall under the "Please don't post shallow dismissals" guideline.
https://simonwillison.net/2026/Sep/18/the-creative-spirit-of...
I don't believe it is that important, but I guess you might've just hit some people's pet peeve with that. If anything, it goes to show as to why skeletons are popular on modern sites
No moat and competition is good for consumers though.
But there is always a use case for frontier models, even if they’re quite a bit more expensive. The set of things you can profitably do with better intelligence than everyone else is unbounded.
So yes, the number of tokens that get pushed through commodity models for very cheap will continue to grow, but so will the tokens for expensive frontier models. We’ll never run out of things to do with the latest geniuses who are twice as smart as last year’s geniuses.
The moat is the engine that builds and sells the latest geniuses, especially the compute. And no one will have more compute than American labs for the next 5-10 years.
I picked double out of a hat, obviously there’s some intelligence gap that’s too small for a price premium beyond a certain point, but it seems like OpenAI and Anthropic’s revenue keeps climbing, so I don’t think we’ve hit it.
in any case, I've been using open and closed models since sonnet 4, i remember when the best I could get was qwen 3 480b coder, you can definitely feel the gap closing going from that and GLM 4.5, to GLM 5.3, DeepSeek Flash V4.1, Kimi K3 etc, it's reached the point where i wish I had V4.1 at work, it's faster and bullshits me less when I use it in my personal projects. And I have unlimited access to fable 5.1
While this is trivially true, the question is if these remaining use cases that separate frontier models from non-frontier models bring in enough revenue to offset the massive spending of the frontier labs.
Which is why while HN is sleeping and thinking Google and MS are done in the space, Copilot and Gemini are winning most of the non-tech: it's just an extension of their current teams/cloud offerings that companies were already using.
I have asked most of my non-tech friends: chemicals, banking, insurance, consultancy. They are all on Gemini or Copilot.
In my opinion, concluding that Copilot is "winning" right now is like concluding that Internet Explorer was winning while Chrome was still young. It may be technically true, but it's easy to see the writing on the wall -- that it's only winning because it is the default option available at most companies.
I have never met a satisfied Copilot user, and as a consultant I met a lot of Copilot users. It's hard to believe how totally and completely Microsoft is shitting the bed. As you point out, they have the enterprise integrations. Why can't the build a good product with them? Generational blunder, in my opinion.
we would be better off using linux all together, but here we are. Lets see maybe every company will have their own little instance of a local LLM in their basement in the future, when the AI-bubble pops and server-racks become cheap
If we mean Gemini the gemini.google.com, or gemini built into sheets or other Google surfaces, then people in these orgs might be getting a very different experience of what LLMs are and what they're capable of.
Gemini the web/app is basically where ChatGPT was on release day, maybe even worse than that. It's laughably bad. It forgets the context between turns, it has dynamic tool injection which doesn't get injected half of the time, which causes model to hallucinate tool calls.
The new Spark is much better, but I don't think most people are using that.
but cost is a thing, if i use the LLM gateway for a chatGPT-5.4 i am quickly (3 days of very light use) on 50$ in token. and everyone and their intern are using copilot. As azure has the guarantee that company data will not leak they are taking the bait.
The fear of secrets landing in the training data is just too big, and they are all microsoft 365 customers already. And lastly Management seeing that copilot can autonomously edit an excel file for them already counts as mindblowing... 0_0
Some features of the release I like:
- Demonstration of diverse tasks, such as using a DAW
- Graphs from various benchmarks and price ranges
- Real world use of the model in scientific environments
Interesting but not surprising trend across the board seems to be, the flash models seems to have caught up with the pro-sized models of H1'26. No surprise all labs are rushing to bigger models.
EDIT: Wow, took a detailed look at the benchmarks. Mimo 2.6 pro, the 1T model leads Kimi K3, a 2.8T param model in 14 out of 15 benchmarks (and the last one is near tie)!! Good to see they also kept the price the same, and landed in the greenest quardrant of the intelligence vs speed of AA.
Averages ~25-35tok/s which isn't bad for a first attempt.
https://aibenchy.com/compare/x-ai-grok-4-7-medium/xiaomi-mim...
Also when I was using it, I managed to do quite a bit on a few bucks worth of OpenRouter credits. Not sure how well it keeps up in the modern world against things like Luna, but I hope it remains competitive
[1] https://sebastianraschka.com/llm-architecture-gallery/per-la...
[2]: See DS 4.1-Flash and Qwen-3.8-Next.
It’s an NVFP4 quant, but it fits, and is surprisingly capable.
(or is it somewhere else)
This one!
I'd recommend pointing your agent at it (after installing sparkrun), and asking it to research the absolute latest in TP=1 Flash-Next - mine grabbed particular vLLM nightlies and mods to improve performance, and it was well worth it.
I have a watchful eye on the diffusion ~ Jev/Kev PR
I'm so tempted to buy a second one...
I'm running embedding, reranking, and policy tuned models too, and a Jev/Kev when that's landed. Flash Next is not a substitute for those
I have OpenCode/Fireworks to access big models
Qwen 3.8 Flash Next (what I'm running basically entirely now) sees 30 / 35.0 / 45 tk/s for prose, analysis and code respectively for actual use (not short context benchmarking) with Pi. Thinking blocks are ~35tk/s or so.
The GB10 having so much compute is great for prefill too, 2000-3000/s for 14k to 64k token prompts (cold cache too) in the quick benchmark I did. 3500tk/s for warm cache which is nice :)
When I accidentally streamed my ngrams over the 2.5Gb/s network, it cut all the throughput down in half basically. Especially notable for the time-to-first-token, which is what clued me in that I'd messed up somehow!
For Qwen 3.8 27B, I got it up to a consistent 20tk-25tk/s but 27B thinks so much that it was honestly too painful: Flash Next is as smart, as useful, but much faster for real agentic dev usage IMO
Laguna S 2.1 saw similar numbers to Flash Next if I remember right, but their latest updates means it doesn't quite fit a GB10 128GB anymore at full context which is a shame.
Note: these are all NVFP4 quants (usually a dynamic one where some tensor layers are left at full precision though)
I want to see about fine-tuning these models a bit on the GB10 to tame that over thinking and some other behaviors (like using tools I don't use)
qwen 3.8 seems to have been trained with some `rkt` that messes with tool outputs to "save tokens"
However, this is for relatively low-end systems, with a couple of fast SSDs providing around 20 GB/s throughput (or with a few more, but connected through relatively slow Thunderbolt, for a similar total throughput).
If you use 16-lane PCIe add-on cards with 4 M.2 slots for SSDs and a total throughput of 50 to 60 GB/s, you can quadruple the previous speed in a desktop PC where you use the GPU PCIe slot for SSDs (a fast CPU, e.g. an AMD 9950X, would be alone fast enough for inference limited by SSD throughput, so a discrete GPU would not be required).
If you have a server/workstation motherboard, e.g. with 6 16-lane PCIe slots, you might gain another factor of 4 in the inference speed, so one might reach around 15 tokens per second for a very big model, but the cost would also be high, with so many SSDs, even if at that number of SSDs each SSD could be the smallest that can be found with a PCIe 5.0 interface.
Flash, like pretty much every solid state storage technology, can scale its output bandwidth to ridiculous levels limited pretty much only by the readout circuitry. There may be a price to pay in power consumption, though.
Close but yet so far. Nobody has released a Q3 that fits so far. I think 95-115gb is the sweet spot
I’ll be trying these models out and may end up switching my subscriptions if this craziness continues
What works for me when I need cyber from Fable/GPT is switching the model after cyber refusal to DS flash for one turn, then switch it back to Fable/GPT. This works most of the time.
It's because offpeak electricity is cheaper?
Funnily it's perfect if you are in the Pacific Time Zone because you can use it daytime 9am to 5pm
And as these models get better the pace of training is quickly speeding up too.
This doesn't bode particularly well for anthropic/OAI after they go public.
with time, it became commoditized, people now have unlimited plans, and the money is made by the applications that sit on top (token generation is increasingly undifferentiated low-level infra)
This is not to say there has not been significant innovation in the time since, but it's a low margin business (tokens look to be headed this way)
They are banking on the application layer and accumulated business and end user context. They have to quickly make that systems integrated value out weigh the model choice price value in the broader market
Unknown if they will be able to pull that off.
https://artificialanalysis.ai/models/mimo-v2-6-pro#intellige...
That's pretty fast; I think I'll try it: https://openrouter.ai/xiaomi/mimo-v2.6-flash
One concern I have is that they allegedly do not discount cached tokens: https://www.reddit.com/r/opencodeCLI/comments/1t37dz3/xiaomi...
Can anyone comment?
16 minutes per test, which is a lot for simple questions...
Sometimes they fail because they reason more than their max context window without giving an answer, that's odd too.
With some models you can find hosting companies based in the EU or US, but then you don't know how they're quantizing the models, so you're not sure about the actual output quality.
How are people actually using this? Or are people just experimenting with side projects?
Hope they bring MiMo for tests.
[1] https://platform.xiaomimimo.com?ref=UKV2FC (invite link = 10% off)
And how's the performance?
And can you choose the server or does it just use your geo to determine? e.g. can I as a Canadian pick the European infra?
Twice, trice or quadrix(tm) are just approximations, we have already paid with:
- more expensive electronics
- less work
- all the retirement money put into gpus
- all the "fair use" of all the books, all the images,
- and then taxes to bail them?Design: https://image.non.io/78795662-8bfc-4e14-8d72-3738392aa6b3.we...
MiMo 2.6 Pro Ultraspeed (36min): https://html.non.io/annui-mimo/
Grok 4.7 (25min): https://html.non.io/Annui-grok/
Astra (19min): https://html.non.io/annui/
Overall this felt like the weakest of the three. Ultraspeed was fast as far as tokens per second goes, but it overthought quite a lot of things resulting it in having one of the longest build times. That overthinking didn't lead to better results either - note the statue with the cropped off arm. It's also the worst implementation of the dynamic lighting effect / displacement effect - the background especially has some significant distortion. Astra was the only one that seemed to understand that displacement should happen less the further something is in the distance.
Here's a vid of all 3 side by side with the source design: https://non.io/video/annui-comparison.mp4
Perhaps instead of giving it a one shot task with vague prompt. I wonder how it can perform with much more detailed and constrained prompt (can you please elaborate more on the level of detailness and ambiguity that the prompt is and where does this model seem to overthink the most?)
Also are there any ways to tame such overthinking of models in general?
I hope that once models start becoming smart enough (I think for me it’s already there) or becoming genuinely the Sota. They then start focusing a lot more on optimizing token usage
KillSwitch-Bench 1.0
Claude Opus 5 66.9
GPT-6 Astra 57.9
Claude Fable 5.1 46.7
MiMo-V2.6-Pro 38.8
Muse Spark 1.3 36.5
- capped per-task budget and time limit
- No internet access
- different harnesses mixed
I would argue that this benchmark is uniquely suited to how most people use LLMs because it actually tests common harnesses and it is a true coding benchmark for a language that is unseen, thus testing the LLMs actual ability to understand nuance and learn.
Internet access is restricted to ensure that over time models cannot just look up the source code of KillSwitch, which would allow them to cheat.
In my experience models just don't take forever to mark tasks as done.
For my coding usage I don't set time limit so benchmarks that do so are providing me less interesting use cases.
With that said, I can understand having budget limits for expensive LLMs for those of us that don't have infinite VC money.
As for no internet access, the issue is that for most problems we do want Llms to be able to search docs, API specs, GitHub issues, etc. So not allowing that usually just favours larger LLMs that were able to memorize more data, not necessarily smarter ones when both have internet access.
But at least I can run Chinese models locally, and strip a lot of that censorship/refusal.
The difference is China has a good reason to. China doesn't look appealing because they're more moral than anyone else, but what they have going for them is that they still behave like a rational actor. At least their behavior is intelligible in terms of their own interests. The world can deal with a long term selfish superpower but not an unhinged one
I don't think there's a person in China that has as much of a seething hatred for America's 'allies' in Europe as J.D. Vance or half of the American techbro commentariat does
And later they further cut Sol and Terra pricing by 20% (maybe only in the API) and Luna by 80%.
In fact Luna still outperformed DeepSeek Flash 4.1 in cost per task on Artificial Analysis when I last checked.
However, Luna is slightly less intelligent. I have a feeling that it's pretty dumb and prone to hallucination unless running at xhigh or max effort, where it somehow manages to work quite well.
I did not personally test the open weight models beyond the old Qwen 3.6 27B, which produced unusably bad results for me.
The competition is great, and I hope Chinese models will continue to force leading US labs to offer models at a low price point.
That said, I don't think the Chinese labs have anything over OpenAI and Anthropic when it comes to capability or efficiency - I have no reason not to believe the US labs have even lower cost to serve the models.
Do also remember China is this far in the AI race despite all chip restrictions from America. If they were in equal standards I truly think Chinese models would have long surpassed American ones. Also would like to remind how Anthropic CEO is being hostile and blaming Chinese models with distilling meanwhile their own models claimed to be Qwen¹ and their stance against open models is negative² and they still keep blaming China for it.
1- https://news.ycombinator.com/item?id=48671252
2-https://www.anthropic.com/news/position-open-weights-models
Why wouldn't he? If there really was 25,000 accounts breaking ToS any CEO would at minimum be upset. Evidence of Claude distilling qwen would be damning but that a) makes no sense b) doesn't exist afaik.
Is this even true?
I don't trust a single word that comes out of thr people behind Anthropic/OpenAI.
Chinese models kept improving in capability regardless, and are in some ways more impressive than Claude/ChatGPT.
So yeah, I think they are bulshitters. The can create reports and send letter to congress simply because they know if allowed to compete freely the Chinese models will eventually prevail.
Also, very rich of you to mention Chinese firms record on IP theft when Anthropic and OpenAI are companies entirely built on large scale IP theft.
Both can be true
Given the difference in compute, it seems plausible.
However, the researchers at the US labs are surely no less talented, and they have better access to hire talent globally.
They too have to serve their models efficiently at a large scale, and with current capacity constraints this must be a top priority.
What do you think is motivating the Chinese labs, benevolence?
Limitations often lead to creativity to overcome them. The Chinese AI labs have had to focus much more on efficiency so they got good at it. Meanwhile breaking new ground is often harder than replicating it. So even if they had matching compute it's not a given they'd be better.
So you don't have much perspective on things, it seems. Let me introduce you to the GLM 5.2 and then 5.3/5.3 flash series of... "oh, wow, I should have bought some RTX PRO 6000's while they were 'cheap'" stage of progression.
As someone carrying multiple max subscriptions to both claude and codex - primary workhorse is glm 5.3 flash running on rented GPUs for less than a latte/hr.
I also found qwen 3.6 27B nearly useless for my own needs. DS4 flash 0731 and then 4.1 have been nearly as eye opening as glm 5.3 flash, but have their own warts.
Or I guess the other way around, if GLM 5.3 Flash is so good, why Claude and Codex?
OpenAI reduced prices and Anthropic increased weekly usage limits.
As a Max x20 and Pro x20 subscriber, can tell you that it doesn't matter since they continually move the baseline of token use So in practice you feel that you're continually getting less from your subscription.
While it never happened to me in the past, i reached my weekly limit within 3 days using Opus 5. And the Open AI weekly limit essentially is a Claude Max x20 5-hour limit. Not even talking about the baseline in intelligence : on release day Astra was so good that it lead me to move to Pro x20. Now it's dumb af and token use is insane.
Deepseek 4.1 Flash has been a lifeboat for me, finally able to work without being constrained/distracted by limits and with what is in my view even better intelligence than Opus 5 for a fraction of the costs. DS is not messing up my brain with load-bearing pseudo jargon in every sentence. It respects coding guidelines, and completes even the most complex tasks most of the time in one shot.
DS 4.1 had been able to add complex features to my repo without breaking a sweat (330k lines of F# + 4M circa lines of an Angular frontend). Writes very idiomatic F# and respects our guidelines and style perfectly. Just completed an extensive UI/UX research and implementation work.
I am ditching both x20 subs and will only keep a Pro x5 because wife does a lot of design work and needs solid image generation capabilities.
debatable if a turn around is possible before '29
China was never ahead of the USA in AI. So perhaps a more proper analogy is the Moon landing. In real history the side that lost the race never got its mojo back...
I expect this to happen within 12-18 months, the differentiation has shrunk, many models are now sufficiently capable for most tasks
I was simply saying that when (not if) Chinese AI models will pass Americans, it will probably be game over and Americans will never catch up, let alone become leaders again.
Check the names of the researchers in the DeepSeek's latest paper. Full of Chinese names. Check the list of names in Google's paper. A very similar view. Anecdotal, but quite thought-provoking...
https://www.nytimes.com/2026/07/09/science/nobel-winning-us-...
One simple task: I needed an LLM to go through and clean up a few thousand page descriptions and titles in my personal search engine index, where the human web page authors had put in no effort sigh. I did a shoot out between Claude, Luna, GLM 5.3 Flash and Deepseek. Despite the high cost, Claude's descriptions were terrible, and even Opus warned me that the descriptions coming back from Haiku were "generalized, not accurate". I expected I would choose Luna because of price, and occasionally it did have wonderful descriptions (one captured emotion in a way no other model did). But in the end, the GLM 5.3 Flash descriptions were the easiest to read, they flow well while also being accurate & including necessary keywords, and being highly affordable. So it won out. It's a task that is nowhere near frontier, but a task where somehow China is better than frontier.
I just had like four big sessions going today, paid about $8 in tokens. I see no reason to pay more, this is more than I need for intelligence.
My impression of DSv4.1-flash was very positive aside from this. But that was enough for me to stick with GLM-5.3(-flash), which both gave me consistently great results
I was using a vibe coded bare bones harness. I was wondering if this was normal from DSv4.1-flash, or if its my harnesses fault.
So if you use MCP a lot, simplify the params, be more lenient on validation and rework the errors.
It is quite good with shell.
No mcp support. I'll try copying deepseek harness's basic tool call formats as a starting point.
glm 5.3 flash is a tad slower but a bit more capable and way more token efficient.
Source: self hosted tested on rented GB200 node at 8bit.
ds 4.1 is lightning fast though. Also much better in image recognition.
Months ago I switched entirely to use Chinese model. Mostly DeepSeek and MiMo, although I recently started to play with GLM as well.
The models are excellent and in many ways I prefer them to Claude.
I see no difference in terms of capability, but the fact that they are cheap frees me to experiment.
So maybe one day Chinese models will squeeze out the American ones, but today is not that day.
So no, I am not excited about Chinese models ( just because its open weight and not American).
Here on HN was a post few days ago titled like "so you want to use openrouter", there was a benchmark in capabilities between providers which showed some aggressively quantize and basically break models and tool calling.
I am in no way a professional power user, but I frequently suffered from "call fails" (e.g. unclosed tags, broken agent loop, broken thinking blocks), so I had to babysit agent on it's loop. After I blacklisted like 20 providers (I think most broken were Nebius and DigitalOcean) these issues completely went away. I had several agents work on my small tasks for 18+ hours with no issues.
I really think they have drunk too much of their own kool aid and become completely detached from what the market wants.
I'm just not going to build long term infra that depends on something that another person can and will - objectively based on experience - take away from me at some unknown point in the future.
The biggest benefit of open models is they keep all the other players honest. The extent to which they feel they can dictate terms is directly set by the threshold where they feel people will take the trade to run open models instead.
Staggeringly low for a frontier model.
Im a bit lazy and only use the free models different companies host and the biggest difference i see is that some models (Gemini, OpenAI) get progressively stupid in long chats. You end up having to start a new session every oncr in a while. Or they get really hung up on a theme and cant shift to a new topic.
By contrast, Ive been impressed with Qwen. I have some chats on research and code architecture that have stretched for weeks without any noteable change in quality (though occassionally it seems to "rush" to an answer)
Im just looking at all the listed benchmarks and im unsue which i should be looking at
Just tried 2.6 flash on a really niche topic I specialise in and it has done a really good job. They’ve definitely polluted their training data with claudeslop, but looking past the slop there is a decent model.
So yes, we are getting scammed by American SOTA.
MiMo-V2.6-Flash-310B-A15B roughly GPT-5.6 Luna / Claude 4.9 according to benchmarks MiMo-V2.6-Pro-1.02T-A42B roughly GPT-5.6 Sol / Opus 5 according to benchmarks
Perhaps with IQ2 flash will run on 128G M5?
> we also removed the cyber dataset from the upcoming pro run, since we observed some bad patterns in the rollout logs.
And this in the model card (emphasis mine):
> Aligned RL: Cold start from self-correction — the model reflects on and rewrites its own misaligned turns into grounded next steps. Throughout RL, environment hardening, adversarial screening, and verifier cross-checks keep the loop honest against reward hacking.
Environment hardening during the RL runs. Uh oh, did someone start making a few too many paperclips?
Gotta love a capable open model. BUT, how can they just casually throw in that they're actively exploring RSI as if it's just another technique? Is this not alarming at all?
I deliberately left things wide open and ambiguous to see what it would do. I think with better upfront planning this would be excellent value.
it is reasonable to question why America doesn't have such environment.
so weird to acknowledge someone being on the front edge, but not name it
but now I got my "proof".
They are excellent in marketing, I guess that is something.
It looks like the "Frontier Line" to me, which is also often misinterpreted. frontier does not mean the best models. It means all models that are not strictly dominated, meaning in most cases: Not same price or cheaper and more intelligent.
I personally would like the word frontier to be used with more criterias: Open Weights, per use-case, etc etc. This would make model selection easier, but I understand it’s not an easy thing to do.
- Pareto efficiency/Pareto curves: Basically the convex hull of points along the edge of a graph, indicating the best tradeoff between the axes. This is what the post is talking about.
- Pareto principle: this is the 80/20 rule you're talking about
What you call "frontier line" is also called "Pareto frontier" https://en.wikipedia.org/wiki/Pareto_front
Your description of it is basically correct though