I know we have strong views on what a truly open model is (open weights, open training data, open training code etc.) but I really like how transparent they’ve been about the training of this model.
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!
ignoramous 2 minutes ago [-]
> got lots of clever behind the scene tricks like Google or Deepseek writeups) and benchmark scores
Perhaps it is due to Luo, how similar Xiaomi's tech & GTM approach is to DeepSeek.
earthnail 49 minutes ago [-]
Thanks so much for sharing this. As someone who mostly watches from the sideline, can you share what you can see in this dashboard that someone like me can't see? Is it the metrics themselves that they measure (the metrics tab is absurdly detailed), something in the notices, or something else I missed?
rao-v 9 minutes ago [-]
I might turn this into a blogpost if folks are interested, but my god there is so much clever info in that dashboard.
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 for models to improve with small tweaks to what they already do than from some perfect oracular 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?!
tancop 19 minutes ago [-]
The best thing they did is being open about all the setbacks they had to deal with. They logged every restart with a reason, talked about dropping a cyber dataset after it degraded coding benchmarks. Also published real time training loss, benchmark scores after every checkpoint and running cost estimates.
Really the only thing missing was dataset descriptions, the dashboard only had random IDs like "dataset-zrso". I guess it's their lawyers fault.
verdverm 37 minutes ago [-]
the existence, who else has a live dashboard for the RL late-training?
MangoCoffee 9 minutes ago [-]
maybe this is why Dario want to slow down AI development and all the big AI labs in the USA is singing the same song.
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.
Looking terrible isn't nessesarily a bad thing. The pelican is heavily pre trained now. Having a crappy pelican means you didn't try to juke the stats.
broodbucket 14 minutes ago [-]
Apologies for not taking the time to find it, but there was a post that tried to determine if the pelican was benchmaxxed across a bunch of models by comparing it to other SVGs, and found that it wasn't at all.
handfuloflight 31 minutes ago [-]
How does this translate to coding performance, which is what most of HN cares about (...I assume)?
simonw 29 minutes ago [-]
It means they're good at writing SVGs, in particular SVGs of animals riding modes of transport!
lanyard-textile 25 minutes ago [-]
I only visit HN for the pelicans, personally.
lukewrites 14 minutes ago [-]
Yeah, I thought the "N" was for Nest
Imanari 33 minutes ago [-]
ish… at least we can be sure they don’t benchmaxx the pelicans lol
lwansbrough 57 minutes ago [-]
Anyone else more excited about Chinese models than American models these days? Big thing for me is affordability.
tacomagick 48 minutes ago [-]
Absolutely! Chinese models are both cheaper and more capable in many cases, compared to the American models and their makers continuously fumbling or reducing model capability with each update. Deepseek decreased costs when they released Flash 4.1 you would not see any American company do this, in reverse they would try charge you more.
user43928 39 minutes ago [-]
OpenAI decreased prices with the 5.6 model family.
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.
tacomagick 28 minutes ago [-]
OpenAI had to cut costs because of Anthropic. I also do not trust the benchmarks when it comes to models anymore. I have tried both Claude and OpenAI models and while it is true that the 5.6 series is smarter than Deepseek (at the time i tested it against 4.0) at that price it is still not worth it and sometimes randomly refuses to do tasks or stops midway etc.
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.
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.
goosejuice 12 minutes ago [-]
> Also would like to remind how Anthropic CEO is being hostile and blaming Chinese models with distilling
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.
goosejuice 20 minutes ago [-]
> Deepseek decreased costs when they released Flash 4.1 you would not see any American company do this, in reverse they would try charge you more.
OpenAI reduced prices and Anthropic increased weekly usage limits.
SyneRyder 30 minutes ago [-]
Yep, I'm trending in that direction, and I'm someone with Claude stickers all over my laptop. My main app dev work is still going to Claude, but everything else is going to China even at API rates now.
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.
swingandamiss 54 minutes ago [-]
No, because I'd rather not support our economic and military rivals.
lwansbrough 52 minutes ago [-]
I'm Canadian so this sentiment has little value in 2026 unfortunately.
ActionHank 40 minutes ago [-]
Also, frankly, as a fellow Canadian it's pretty clear that the biggest "rival" the US has right now is itself. Just passed out in the corner puking on itself shouting about all the foreigners who won't talk to it.
16 minutes ago [-]
tancop 10 minutes ago [-]
I'm from Europe and I hate America way more than China now. Used to be about equal but then Trump started extorting Ukraine, threatening their own allies and sending billions to Israel to help with a genocide. I think that exposed America for what it really is.
scottyah 38 minutes ago [-]
[flagged]
lwansbrough 34 minutes ago [-]
Because at present the pedophile US president is making it his mission to molest my country. China, for all its faults (including espionage, which the US is also guilty of) is mostly focused on conducting trade.
verdverm 34 minutes ago [-]
Half of Canada now uses the word 'enemy' when asked for an adjective to describe America or China. We're equivalent in their eyes now because we elected Trump a second time and all that he has said and done in 2.0
cwillu 13 minutes ago [-]
It's closer to a cousin you used to be close with despite some moral failings, but who has now has a substance abuse problem and is lashing out at family and friends.
Not an enemy, just a danger.
verdverm 3 minutes ago [-]
I'm relaying a poll of Canadians, their word choice, not mine
"plurality" would have been accurate rather than "half"
Canadians warming up to China makes me think of Germany becoming increasingly reliant on Russia in the 2010s.
cgio 1 minutes ago [-]
Yes, someone can still blow up a pipe and they look the other way. On the other hand, you can also draw parallels to themselves becoming increasingly reliant on US vs UK in the past.
bigyabai 2 minutes ago [-]
Mind you, Canada was a reliable defense and trade partner before the US' arbitrary trade war and annexation threats.
Freedom2 53 minutes ago [-]
Agreed, and also because I support freedom of speech!
girvo 45 minutes ago [-]
Neither the US nor the Chinese companies are on your side then. They both censor, just different topics.
But at least I can run Chinese models locally, and strip a lot of that censorship/refusal.
verdverm 35 minutes ago [-]
I have a contrarian opinion that China passing America in Ai is the Sputnik moment we need to leave the hubris behind and get our mojo back
Those this mean they've fine-tuned this Qwen 3.5 9B on output from the V2.6 model?
mydreamof 53 minutes ago [-]
It is a 9B agentic model developed by Xiaomi MiMo through supervised fine-tuning of Qwen3.5-9B on MiMo-generated data
simonedepertis 52 minutes ago [-]
[flagged]
verdverm 1 hours ago [-]
curious why the HF pill (on the right) always has inaccurate values
stymaar 57 minutes ago [-]
I noticed the same, and I wonder as well.
verdverm 49 minutes ago [-]
I suspect they are calculating something in the weights or config, I see it pretty consistently with quants
segmondy 28 minutes ago [-]
more like 500B in FP8
nemothekid 1 hours ago [-]
Looking at the frontend design examples; why do these models seem to love the "01 - UPPERCASE TEXT" motif. It's everywhere now (see https://try.cloudflare.com/, which has '01 · QUICK TUNNELS', but no "02" anywhere).
danvayn 1 hours ago [-]
My guess is that by function they break down frontend sections or components into pieces and I believe document things for themselves on some level, or purposely are verbose in this way. It is probably also shaped by users and existing web patterns. They probably get reinforced by models the more common they become.
sandblast 1 hours ago [-]
Nice catch!
pphysch 22 minutes ago [-]
The extraneous small-caps labels are one of the main idiosyncrasies of AI generated markup. I wonder how much of this is a "scaffolding" technique to help the model build stable designs. But was it reinforced in RLHF or an emergent behavior of the models?
user43928 56 minutes ago [-]
I don't trust any of the benchmarks where Opus 5 surpasses Astra or Fable 5.1.
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
ExploitGym
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
DeepSWE v1.1
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 19.0
mokre 54 minutes ago [-]
Maybe you should not trust any of the benchmarks!
varispeed 29 minutes ago [-]
They match my experience. Astra and Fable I rate below Sonnet. They are incredibly poor. They were excellent for a couple of days after release and then plummeted.
Maybe I am being routed to more quantised versions or less capable models with system prompt to fake Astra or Fable.
4 minutes ago [-]
vatsachak 1 hours ago [-]
Wow, the chinese labs are getting good at advertising model releases. The moat is thin.
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
syntaxing 1 hours ago [-]
All these new models are such tease for us folks with 128GB of shared memory. Buying another unit now to expand to 256GB is a mortgage payment but it’s getting tempting…
there are many tasks (increasingly more each day) where small models are more than enough
brcmthrowaway 1 hours ago [-]
Is there a gamechanger around the corner to reduce DRAM requirements?
zozbot234 52 minutes ago [-]
You could always stream from SSD storage. Especially effective if you get a cheap old-gen HEDT with lots of PCIe slots to add NVMe storage to and reasonable overall PCIe bandwidth.
jkingsman 13 minutes ago [-]
That nearly certainly boots you to secs-per-tok land (as opposed to tok/s). Plausible if you are willing to wait hours to days for responses for simple testing, but not (debatably) "usable".
this is to offload VRAM to DRAM (for GP comment), and makes no difference for URAM
zozbot234 47 minutes ago [-]
You can definitely offload n-gram embeddings to storage; they're very sparsely used (only a few KB fetched per token) so this is quite effective. Loading to DRAM only becomes necessary if they are a bottleneck to overall performance (which might happen if you're doing very wide batches and everything else uses super fast VRAM/HBM).
verdverm 44 minutes ago [-]
I was looking at the qwen-next-flash, and the weights would fill my OEM Spark on their own, before the n-gram. I'm unclear if offloading to disk can work here, is that what you are implying is possible?!
girvo 43 minutes ago [-]
Check out eugr’s TP=1 sparkrun recipe :)
It’s an NVFP4 quant, but it fits, and is surprisingly capable.
verdverm 42 minutes ago [-]
do you have a HF link? HF search is not uncovering it for me
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.
verdverm 11 minutes ago [-]
I have a quirky vLLM on k8s on 2x OEM sparks setup with 9 models available to me. I'm not keen to run nightly vLLM, too many issues with it in the past. Going the qwen-next path means displacing things I use daily :/
I have a watchful eye on the diffusion ~ Jev/Kev PR
For what it's worth, Flash Next outperforms every other model that is available to us on the GB10 in all of my testing; though if you have two sparks then the TP=2 version is even better and easier (I don't think you'll need the nightly for that at all, just use the recipe)
Nah, I’m streaming ngrams off NVMe on my Spark-alike right now. Works surprisingly well (except for when I accidentally bottlenecked it through my NAS)
jkingsman 10 minutes ago [-]
What kind of throughput do you see on what models?
verdverm 43 minutes ago [-]
interesting, peer comment seems to indicate this is a possibility as well, will have to take a deeper look
petu 42 minutes ago [-]
n-grams can be kept on SSD, no need to hold them in any kind of RAM (at least w/o batching)
stymaar 60 minutes ago [-]
Am I missing a joke? WTF is URAM?
verdverm 58 minutes ago [-]
unified memory, not sure if anyone uses URAM, I human hallucinated it
Averages ~25-35tok/s which isn't bad for a first attempt.
1 hours ago [-]
thrownawaysz 42 minutes ago [-]
>Night 0.8x Usage, 00:00-08:00 -UTC+8
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
eriquesito 23 minutes ago [-]
Funny that all but one video has audio, the house 3D model one, where you can hear (what I assume are) Xiaomi's engineers talking about who knows what.
ddxv 1 hours ago [-]
This looks great in terms of cost and capabilities, truly pushing the frontier forward in terms of open weight light weight models.
MisterMunchkin 1 hours ago [-]
I really liked MiMo 2.5, it was really affordable and actually had vision, unlike DeepSeek. (DeepSeek has only recently added it)
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.
omani 1 hours ago [-]
how do you recognize "claudeslop"?
Bluestein 56 minutes ago [-]
It's an honest, load-bearing, simple thing.-
DanMcInerney 1 hours ago [-]
This is a big week. Probably getting next OpenAI and Anthro models, Grok 4.7, Mimo, etc. These open source model releases are why I can't take the "slow down" crowd seriously. I pitted older Mimo, qwen, step, gpt-oss, and other models against each other playing games like Werewolf and Sketch.io-like games where I let them talk shit while they played against each other. Mimo was by far pareto frontier of game-playing for the models that were <$0.15/m input tokens on OpenRouter. Qwen was pareto frontier in the shit talking game though. Qwen's hilarious. https://www.tiktok.com/@clankerfights/video/7642862917582425...
algoth1 1 hours ago [-]
Finally a lab that doesn't cheat on the charts
bertili 1 hours ago [-]
They mixed up DeepSeek 4.1 Flash with something else on this page, possibly DeepSeek 4.1 Flash means Gemini 3.8 Flash.
varispeed 31 minutes ago [-]
These benchmark are useless as they don't say whether they were done before or after Fable and Astra got nerfed.
gigatexal 31 minutes ago [-]
Leaning into what it cost to train is hilarious and an obvious shot at US frontier labs spending tens to hundreds of millions or more to train their models.
alfalfasprout 36 minutes ago [-]
The moat for OAI and anthropic seems to be very quickly shrinking. Chinese labs are now using RSI-like approaches and even without resorting to heavy distillation they're catching up in a couple of months vs. what would have been 6-12 months a year prior.
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.
verdverm 29 minutes ago [-]
token vendors are headed to the same place mobile data vendors went, this is good for everyone but those who thought they could maintain exorbitant prices
NooneAtAll3 1 hours ago [-]
does anyone know what unnamed model is on paretto frontier picture right between MiMo 2.5 and 2.6?
so weird to acknowledge someone being on the front edge, but not name it
AnodicElegy 1 hours ago [-]
Pretty sure that's Luna xhigh.
spwa4 1 hours ago [-]
As for the stats that everyone wants:
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?
16t96 37 minutes ago [-]
[flagged]
nlcs 1 hours ago [-]
[flagged]
nofpu 43 minutes ago [-]
[dead]
unpopularopp 1 hours ago [-]
[flagged]
InsideOutSanta 1 hours ago [-]
It's funny, I have the exact opposite reaction. This is probably misguided on my part, but Xiaomi is one of the very few major tech companies that I don't have an immediate strong negative reaction to. Everything I've bought from them, from robot vacuum to mobile phone, has been reasonably well designed, didn't break, and was priced fairly. I also think their car looks badass.
I'm sure they're doing all kinds of terrible things, like all major companies. I just can't help but like them. Also, this model looks great, and I'll give their subscription a shot next month.
A_D_E_P_T 30 minutes ago [-]
I must second this.
I'm in Europe, and here the options for home appliances are usually German (e.g. Philips), Balkan (e.g. Gorenje), or Xiaomi. Xiaomi is the best by far, and it's honestly not even close.
Their home appliances are so rock-solid that they actually still surprise me. I've gone from, e.g., having to replace electric water kettles every six months to buying one from Xiaomi and never replacing it. (Nigh on three years now.)
I really have a very positive impression of them.
bel8 39 minutes ago [-]
Which model did you have?
I ask because my wife has the 15T and the camera is better than my iPhone 17 Pro. And while toying around with it I didn't notice any bloat.
Plus hers support native split screen which I kinda need to multitask on the go.
I'm so pissed at how bad Siri is compared to her android phone that I'm thinking about selling the iPhone to get a Huawei Pura Ultra.
algoth1 1 hours ago [-]
I still have a xiaomi mi 11 lite, my wife has a 15t. The cameras are the best for the price. The way they chove ads down your throat at every opportunity should be illegal though
platinumrad 1 hours ago [-]
It's a big company, like Microsoft or Google. Some of their products are good and some are bad.
verdverm 1 hours ago [-]
ironic to this thread, I have less bloatware and ads since I switched from Verzion to Pixel on Fi (many years ago)
Curious if Verizon / ATT still force apps on your phone, eg. NFL and Amazon apps, Fi service is subpar
omani 1 hours ago [-]
ah, would you look at that. I was wondering why mimo 2.5 became "dumber" the last weeks. I was speculating they are probably about to release a new version of the model. because the model really acted out a lot. especially the last two weeks. dont know, was just a feeling, highly speculative.
but now I got my "proof".
sandblast 1 hours ago [-]
I guess that would only be possible if your provider was Xiaomi itself?
omani 1 hours ago [-]
yes. I use opencode and opencode uses Xiaomi as a provider.
jwpapi 57 minutes ago [-]
In the chart they use "Pareto Line", which I think is wrong. Pareto is 20% effort leading to 80% results. Which could be interpreted as models costing 20% having 80% of peak intelligence, but that’s not what it looks like to me.
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.
abound 51 minutes ago [-]
There are two (or more) concepts named after the same person:
- 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
Your description of it is basically correct though
shmolyneaux 52 minutes ago [-]
This is the Pareto Front [1], rather than the Pareto principle. It's the idea that anything that's more intelligent is more expensive and anything that's less expensive is less intelligent.
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!
Xiaomi MiMo is led by Luo Fuli, a former Alibaba & DeepSeek employee: https://newsen.pku.edu.cn/news_events/news/people/15385.html (https://archive.vn/I8Pmu) / (https://archive.vn/sb3B6)
Perhaps it is due to Luo, how similar Xiaomi's tech & GTM approach is to DeepSeek.
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 for models to improve with small tweaks to what they already do than from some perfect oracular 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?!
Really the only thing missing was dataset descriptions, the dashboard only had random IDs like "dataset-zrso". I guess it's their lawyers fault.
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.
Pelicans for Pro: https://tools.simonwillison.net/markdown-svg-renderer?url=ht...
Qwen3.8-27b pelican was amazing on Mac.
https://www.nudgehost.com/dpjn3uwe
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
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.
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.
OpenAI reduced prices and Anthropic increased weekly usage limits.
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.
Not an enemy, just a danger.
"plurality" would have been accurate rather than "half"
https://www.commondreams.org/news/canadians-us-enemy-poll
But at least I can run Chinese models locally, and strip a lot of that censorship/refusal.
debatable if a turn around is possible before '29
Pro [2]:, 1.02T total / 42B activated parameters
[1]: https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL
[2]: https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL
https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B
Maybe Terminal Bench 4.0 and ExploitGym are reasonable.
Terminal Bench 4.0
ExploitGym DeepSWE v1.1Maybe I am being routed to more quantised versions or less capable models with system prompt to fake Astra or Fable.
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
[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
https://github.com/vllm-project/vllm/pull/57250
I'm so tempted to buy a second one...
https://github.com/spark-arena/eugr-recipes
https://github.com/eugr/spark-vllm-docker
Averages ~25-35tok/s which isn't bad for a first attempt.
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
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.
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.
so weird to acknowledge someone being on the front edge, but not name it
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?
I'm sure they're doing all kinds of terrible things, like all major companies. I just can't help but like them. Also, this model looks great, and I'll give their subscription a shot next month.
I'm in Europe, and here the options for home appliances are usually German (e.g. Philips), Balkan (e.g. Gorenje), or Xiaomi. Xiaomi is the best by far, and it's honestly not even close.
Their home appliances are so rock-solid that they actually still surprise me. I've gone from, e.g., having to replace electric water kettles every six months to buying one from Xiaomi and never replacing it. (Nigh on three years now.)
I really have a very positive impression of them.
I ask because my wife has the 15T and the camera is better than my iPhone 17 Pro. And while toying around with it I didn't notice any bloat.
Plus hers support native split screen which I kinda need to multitask on the go.
I'm so pissed at how bad Siri is compared to her android phone that I'm thinking about selling the iPhone to get a Huawei Pura Ultra.
Curious if Verizon / ATT still force apps on your phone, eg. NFL and Amazon apps, Fi service is subpar
but now I got my "proof".
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
[1]: https://en.wikipedia.org/wiki/Pareto_front