MAI-Thinking-1

(microsoft.ai)

144 points | by LER0ever 3 hours ago

17 comments

  • keeda 2 hours ago
    > Second, clean data. MAI-Thinking-1 was trained on clean and appropriately licensed data, with AI-generated content excluded from pre-training. This matters for quality, provenance, and control. If we cannot account for what shaped a model, we cannot fully understand its behavior or credibly improve it.

    Shots fired?

    It would be interesting to see how far "clean data" can go on the scaling laws.

    • foresterre 42 minutes ago
      I would really like to see what "appropriately licensed data" means. Cannot imagine they didn't copy all open repo's on GitHub, and can't imagine they asked for permission, or are reproducing license texts from these repo's now. It sounds hand wavy.

      P.S. A fairly basic website otherwise, but it unfortunately seems to be hacking scroll for no good reason.

      • stingraycharles 38 minutes ago
        I assume they took the actual repos’ licenses info account. I don’t understand why they should ask for permission when the license would already allow for it.
        • rocqua 33 minutes ago
          Which licenses allow usage for training? MIT, BSD, etc likely do. But I would expect it gets weird for all the various copyleft licences.
          • cortesoft 6 minutes ago
            Why would it get weird for those?
            • rzmmm 0 minutes ago
              Theoretically it mandates that derivative works use same license but it's unclear if that applies to LLM outputs.
    • andai 0 minutes ago
      Interesting. Wasn't their previous attempt (Phi) trained mostly on synthetic data?
    • supermdguy 29 minutes ago
      It's interesting because their last model series (Phi) was based around the thesis that high-quality synthetic data is better than a large pre-training corpus.
    • vdfs 1 hour ago
      I doubt any lab would say otherwise, they all _claim_ to use licensed data
      • keeda 1 hour ago
        Maybe, but Microsoft, through their partnership with OpenAI, is already involved in major copyright lawsuits. That is probably a driving force for this move, actually... I doubt they would want to tempt fate while those lawsuits are on-going.
    • swalsh 17 minutes ago
      I'd assume it's not up to par with Qwen-3.5 then, which has been distilling Claude, and the quality of the model is probably a direct result of that.
    • onlyrealcuzzo 2 hours ago
      I'm interested how much "Clean Data" is synthetic data from "unclean" models...
      • bicx 1 hour ago
        So, laundered data?
      • ertgbnm 1 hour ago
        > with AI-generated content excluded from pre-training.

        > without distillation from third-party models

        sounds like zero unless they are lying.

        • zamalek 1 hour ago
          > with AI-generated content excluded from pre-training.

          Though this is largely impossible these days, unless they pre-trained on pre-AI era data.

        • saghm 43 minutes ago
          "how many of those shapes are rectangles?" "sounds like zero unless they are squares"

          Adding "unless" to a statement makes it vacuous if the latter clause is weaker than the first clause. I find it hard to believe that a company willing to violate licenses would have scruples about lying about it.

          • rocqua 28 minutes ago
            Not vacuous, but tautological. Which is different, because tautologies can actually be quite directly informative. Whereas vacuous truths tend to be oblique.

            Also, “Microsoft is lying” is not a logically stronger statement, because they might be lying about something other than whether they distilled or trained on AI output.

          • chongli 34 minutes ago
            Adding "unless" to a statement makes it vacuous if the latter clause is weaker than the first clause

            I think that's the point. "How do I say they're lying without outright saying they're lying?"

            It's a common rhetorical trick.

      • xavriley 2 hours ago
        “ We trained it from the ground up on enterprise grade, clean and commercially licensed data, without distillation from third-party models.”
        • azinman2 1 hour ago
          aka all of GitHub OSS
          • ChicagoDave 45 minutes ago
            Yeah this is exactly what I was thinking.
  • __natty__ 45 minutes ago
    It's good there is a new player on the market, I take benchmark tables with a grain of salt, however. Speaking about model presentation it's funny to see how clearly their website is inspired by other AI company blogs with extra innovation of hijacked scrollbar.
  • Alifatisk 48 minutes ago
    > MAI-Thinking-1 is built with enterprise readiness in mind. It supports long context with a 256k token window

    Isn’t 1M becoming the norm?

    • vb-8448 30 minutes ago
      1M it's only marketing, in my experience above 150k quality noticeable drops.

      Claude code will suggest you to start a new session or compact if you go above 100k.

    • stingraycharles 37 minutes ago
      Yes it is, but I can imagine that they want to start out a bit smaller to see how well things scale, and/or did not yet have the time to work on optimizing for the large context windows.
      • droidjj 35 minutes ago
        I struggle to get quality results from the frontier models at contexts > 256k anyway.
        • stingraycharles 4 minutes ago
          Yup, same experience, it’s because the attention basically has exponential complexity. So at large context windows, they need to compress the attention (eg group multiple tokens together), when then leads to loss in accuracy.

          It’s almost always better to keep your context windows small.

  • pixeldash928 3 hours ago
    Looks like the OAI divergence is finally taking place. Seems like the comparisons are mainly with Opus 4.6 and GPT 5.4 though. Still, exciting to see a new frontier player.
    • i_have_an_idea 1 hour ago
      Is it a frontier player though, or perhaps a new benchmaxxed model? People were saying similar things about Grok but it ultimately amounted to little.
      • wasabi991011 1 hour ago
        "preferred by humans over Sonnet 4.6" makes it pretty clearly not benchmaxxed though.

        At least when you define benchmaxxed as "good in benchmarks but not human preference".

    • dude250711 38 minutes ago
      Post 4.6 Anthropic models do not exactly have a stellar reputation, so that choice is smart.
  • jampekka 24 minutes ago
    The benchmarks are a bit of a disaster? It's at about DeepSeek V3.2 level, but with about 50% more parameters. Loses handily to the also smaller GLM-5.1, and even worse to the similarly sized Kimi K2.6.
    • sailingparrot 12 minutes ago
      Yes and no. Yes from a user PoV, I don't really see a great reason to use this other than for enterprises that care about using a model not trained on copyrighted data (not sure what the market really is for this anymore, feels like this concern has been forgotten by most customers.)

      From a strategic PoV for MS, all the model you cited are distilling GPT/Claude/Gemini and wouldn't be anywhere as good as they are without this distillation, which in turn means you are dependent on OAI/Anthropic/G first shipping a good model to generate data for your training. This MAI model is trained from scratch with no synthetic data or distillation. So in term of benchmark its obviously much harder to get strong score and thus not a disaster if they can keep on improving.

    • usef- 8 minutes ago
      They claim to not be training to the benchmarks at all. It'll be interesting to see how it stacks up in actual use.
  • Centigonal 43 minutes ago
    > MAI-Thinking-1 is a 35B-active, ~1T-total parameters, sparse Mixture of Experts model, a smaller inference footprint than much larger models.

    This seemingly nonsensical sentence (of course this will have a smaller inference footprint than larger models) suggests this model's competitors have larger inference footprints and total parameter sizes.

  • BeetleB 1 hour ago
    Based on the first table, why would I pick this over GLM?
    • missedthecue 1 hour ago
      Because your employer might make you exclusively use enterprise copilot.
      • BeetleB 33 minutes ago
        As long as my employer is footing the bill, fine.

        For personal stuff this release is not noteworthy.

  • lordmauve 1 hour ago
    We need to see DeepSWE scores. SWE Bench Pro is junk.
  • kaicianflone 37 minutes ago
    Is that a pretext zoom effect when changing screen dimensions? Very cool.
  • wmf 1 hour ago
    At least there shouldn't be any complaints about benchmaxing this time.
    • i_have_an_idea 1 hour ago
      Just because it is performing rather poorly by comparison, it doesn’t mean it isn’t benchmaxxed. It can still be worse than it appears.
      • wasabi991011 1 hour ago
        It isn't benchmaxxed because they are using human preference as an evaluation.
  • hartator 1 hour ago
    I like it so much when a website hijacks the way my scroll works. This is truly innovative.
  • gigatexal 32 minutes ago
    Anyone believing those benchmark numbers from a 35B model?
    • jeffdn 29 minutes ago
      It says right at the top, 35B active, 1T total.
  • kstenerud 1 hour ago
    They've hijacked scrolling. They've hijacked the spacebar. It flickers like crazy when I try to move through the article. Trying to get through it is an exercise in madness.
    • t-sauer 1 hour ago
      I do not understand how scroll hijacking is still a thing. Who thinks this is a better experience?
    • grassfedgeek 1 hour ago
      Even without flicker it is very distracting. Why do people think this is a good idea?
    • AirMax98 1 hour ago
      I normally don't comment on matters of taste like this, but wow this is brutal. It's like someone threw the site in a vat of molasses.
    • aniceperson 1 hour ago
      there is also a gap between the header and the top of the page... they should ask the ai to make it better a few more times...
    • blisstonia 1 hour ago
      I gave up after the first scroll.
  • vcryan 1 hour ago
    It really looks like they used Claude to design this webpage. I guess the color taupe it the marker of good AI today.
  • bossyTeacher 1 hour ago
    7 modes launched. 5 models in the dropdown. Only 4 actually usable :(

    About time Microsoft joined the fray. After the OpenAI divorce, it really looked like Microsoft was going to become another Uber.

    • giancarlostoro 1 hour ago
      They still own 27% of OpenAI, this IPO will feed them a lot of easy cash.
  • simjnd 3 hours ago
    Absolutely disgusting scroll jacking, even when "Accessibility mode" is turned on
    • dang 2 hours ago
      I'm sure most of us agree, but:

      "Please don't complain about tangential annoyances—e.g. article or website formats, name collisions, or back-button breakage. They're too common to be interesting."

      https://news.ycombinator.com/newsguidelines.html

      • simjnd 1 hour ago
        Forgot about this, my bad!