Xiaomi MiMo v2.6

(mimo.xiaomi.com)

183 points | by volf_ 1 hour ago

24 comments

  • rao-v 45 minutes ago
    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!

    • earthnail 15 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?
      • verdverm 3 minutes ago
        the existence, who else has a live dashboard for the RL late-training?
    • kingstnap 12 minutes ago
      [dead]
  • simonw 8 minutes ago
  • stymaar 42 minutes ago
    Flash[1]: 309B total / 15B activated parameters

    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

    • verdverm 26 minutes ago
      • gandreani 23 minutes ago
        Those this mean they've fine-tuned this Qwen 3.5 9B on output from the V2.6 model?
        • mydreamof 19 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 18 minutes ago
          [flagged]
    • verdverm 31 minutes ago
      curious why the HF pill (on the right) always has inaccurate values
      • stymaar 23 minutes ago
        I noticed the same, and I wonder as well.
        • verdverm 15 minutes ago
          I suspect they are calculating something in the weights or config, I see it pretty consistently with quants
  • lwansbrough 23 minutes ago
    Anyone else more excited about Chinese models than American models these days? Big thing for me is affordability.
    • verdverm 1 minute 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

      debatable if a turn around is possible before '29

    • tacomagick 14 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 4 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.

        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.

    • swingandamiss 19 minutes ago
      No, because I'd rather not support our economic and military rivals.
      • lwansbrough 17 minutes ago
        I'm Canadian so this sentiment has little value in 2026 unfortunately.
        • scottyah 3 minutes ago
          Because your country is increasingly owned by Chinese?
          • lwansbrough 0 minutes ago
            Because at present the pedophile US president is making it his mission to molest my country. China, for all its faults and espionage (which the US is also guilty of) is just trying to conduct trade.
          • verdverm 0 minutes ago
            half of Canada now uses the word 'enemy' when asked for an adjective to describe America or China
        • ActionHank 6 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.
      • Freedom2 18 minutes ago
        Agreed, and also because I support freedom of speech!
        • girvo 11 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.

  • nemothekid 47 minutes 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 36 minutes 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 40 minutes ago
      Nice catch!
  • vatsachak 52 minutes 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

  • user43928 22 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 20 minutes ago
      Maybe you should not trust any of the benchmarks!
  • syntaxing 41 minutes 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…
    • verdverm 25 minutes ago
    • brcmthrowaway 38 minutes ago
      Is there a gamechanger around the corner to reduce DRAM requirements?
      • zozbot234 18 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.
      • stymaar 32 minutes ago
        n-gram per-layer embeddings[1][2] might be it.

        [1] https://sebastianraschka.com/llm-architecture-gallery/per-la...

        [2]: See DS 4.1-Flash and Qwen-3.8-Next.

        • verdverm 29 minutes ago
          this is to offload VRAM to DRAM (for GP comment), and makes no difference for URAM
          • zozbot234 13 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 10 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 9 minutes ago
                Check out eugr’s TP=1 sparkrun recipe :)

                It’s an NVFP4 quant, but it fits, and is surprisingly capable.

                • verdverm 7 minutes ago
                  do you have a HF link? HF search is not uncovering it for me

                  (or is it somewhere else)

          • girvo 10 minutes ago
            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)
            • verdverm 9 minutes ago
              interesting, peer comment seems to indicate this is a possibility as well, will have to take a deeper look
          • petu 8 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 26 minutes ago
            Am I missing a joke? WTF is URAM?
            • verdverm 24 minutes ago
              unified memory, not sure if anyone uses URAM, I human hallucinated it
  • thrownawaysz 8 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

  • alfalfasprout 2 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.

  • ddxv 49 minutes ago
    This looks great in terms of cost and capabilities, truly pushing the frontier forward in terms of open weight light weight models.
  • MisterMunchkin 30 minutes 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 27 minutes ago
      how do you recognize "claudeslop"?
      • Bluestein 22 minutes ago
        It's an honest, load-bearing, simple thing.-
  • DanMcInerney 49 minutes 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 55 minutes ago
    Finally a lab that doesn't cheat on the charts
  • bertili 36 minutes ago
    They mixed up DeepSeek 4.1 Flash with something else on this page, possibly DeepSeek 4.1 Flash means Gemini 3.8 Flash.
  • NooneAtAll3 34 minutes 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 26 minutes ago
      Pretty sure that's Luna xhigh.
  • spwa4 32 minutes 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 3 minutes ago
    [flagged]
  • nofpu 9 minutes ago
    [dead]
  • nlcs 51 minutes ago
    [flagged]
  • omani 48 minutes 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 38 minutes ago
      I guess that would only be possible if your provider was Xiaomi itself?
      • omani 30 minutes ago
        yes. I use opencode and opencode uses Xiaomi as a provider.
  • unpopularopp 49 minutes ago
    I've never used worst smartphones than anything from Xiaomi, bloated ad infested borderline malware territory fork of Android. Maybe just me but whenever I see them on HN I just can't think anything good about this company.
    • bel8 5 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.

    • InsideOutSanta 30 minutes 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.

    • algoth1 30 minutes 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 46 minutes ago
      It's a big company, like Microsoft or Google. Some of their products are good and some are bad.
      • verdverm 27 minutes 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

  • jwpapi 23 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.