24 comments

  • mohsen13 hours ago
    Cursor Composer 1 was Qwen and this is Kimi. IDE is based on VSCode. The entire company is build on packaging open source and reselling it.<p>Ollama is also doing this.<p>There is so much money to be made repackaging open source these days.<p>So funny to see Twitter go wild saying &quot;a 50 person team just beat Anthropic&quot; blah blah.
    • miroljub44 minutes ago
      &gt; Cursor Composer 1 was Qwen and this is Kimi. IDE is based on VSCode. The entire company is build on packaging open source and reselling it.<p>The question is, where&#x27;s the outrage? Why are there no headlines &quot;USA steals Chinese tech?&quot; &quot;All USA can do is make a cheap copy of Chinese SOTA models&quot;.<p>&gt; So funny to see Twitter go wild saying &quot;a 50 person team just beat Anthropic&quot; blah blah.<p>Well, if it&#x27;s an American company, then it&#x27;s a noble underdog story. When Chinese do it, they are thieves leeching on the US tech investment.<p>It&#x27;s all so predictable, even the comments here.
      • chzblck21 minutes ago
        because its open source.
    • NitpickLawyer3 hours ago
      &gt; packaging open source and reselling it.<p>It&#x27;s a bit more than that. They have plenty of data to inform any finetunes they make. I don&#x27;t know how much of a moat it will turn out to be in practice, but it&#x27;s something. There&#x27;s a reason every big provider made their own coding harness.
      • pbowyer3 hours ago
        Can anyone enlighten me how having a coding harness when for most customers you say &quot;we won&#x27;t train on your code&quot; helps you do RL? What&#x27;s the data that they rely on? Is it the prompts and their responses?
        • __mharrison__42 minutes ago
          Does &quot;code&quot; include the prompt? Seems like the prompts would be the goldmines. Hook those up to rl an open weight model...
        • rubymamis3 hours ago
          I guess they rely on many people not toggling privacy-mode on?
        • victorbjorklund2 hours ago
          I doubt the majority does that. I bet the majority is using the defaults.
        • doctorpangloss27 minutes ago
          It doesn&#x27;t matter what your privacy setting is, with any savvy vendor. Your data is used to train by paraphrasing it, and the paraphrasing makes it impossible to prove it was your data (it is stored at rest paraphrased). Of course the paraphrasing stores all the salient information, like your goals and guidance to the bot to the answer, even if it has no PII.
    • dmix50 minutes ago
      Cursor’s integration is much deeper than just plugging an LLM into VSCode<p>That said I have a feeling both VSCode and Claude code will catch up to their integration. But neither comes close yet (I say that as someone who mainly uses Claude Code).
      • bearjaws5 minutes ago
        As a command line junkie, what is the main thing Claude Code needs to catch up with cursor?<p>I haven&#x27;t dove into using a LLM in my editor, so I am less familiar with workflows there.
    • rvz1 hour ago
      &gt; Cursor Composer 1 was Qwen...<p>We know Composer 2 is Kimi K2.5 from that tweet. Where is the evidence for Composer 1 being based on Qwen?<p>&gt; So funny to see Twitter go wild saying &quot;a 50 person team just beat Anthropic&quot; blah blah.<p>In this case, it will be the other way round: Anthropic will see Cursor as a competitor AI lab using open weight models for Composor 2 (actually Kimi K2.5) which was allegedly distilled from Opus 4.6, and would be enough for Anthropic to cut off Cursor from using any of models.<p>That&#x27;s where it is going.
    • rubymamis3 hours ago
      Do you know what Qwen model Composer 1.5 used?
  • deaux3 hours ago
    Looks like two Moonshot employees confirmed that it&#x27;s not licensed before Moonshot made the decision to get out of the debate and delete their posts [0][1].<p>[0] <a href="https:&#x2F;&#x2F;chainthink.cn&#x2F;zh-CN&#x2F;news&#x2F;113784276696010804" rel="nofollow">https:&#x2F;&#x2F;chainthink.cn&#x2F;zh-CN&#x2F;news&#x2F;113784276696010804</a> - may have originally been <a href="https:&#x2F;&#x2F;x.com&#x2F;apples_jimmy&#x2F;status&#x2F;2034920082602864990" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;apples_jimmy&#x2F;status&#x2F;2034920082602864990</a><p>[1] <a href="https:&#x2F;&#x2F;pbs.twimg.com&#x2F;media&#x2F;HD2Ky9jW4AAAe0Y?format=jpg&amp;name=medium" rel="nofollow">https:&#x2F;&#x2F;pbs.twimg.com&#x2F;media&#x2F;HD2Ky9jW4AAAe0Y?format=jpg&amp;name=...</a>
    • lmc3 hours ago
      This is on their website...<p>&quot;Is Kimi K2.5 open source?&quot;<p>&quot;Yes, Kimi K2.5 is an open source AI model. Developers and researchers can explore its architecture, build new solutions, and experiment openly. Model weights and code are publicly available on Hugging Face and the official GitHub repository.&quot;<p><a href="https:&#x2F;&#x2F;www.kimi.com&#x2F;ai-models&#x2F;kimi-k2-5" rel="nofollow">https:&#x2F;&#x2F;www.kimi.com&#x2F;ai-models&#x2F;kimi-k2-5</a>
      • saidmukhamad3 hours ago
        4th paragraph in license block<p>Our only modification part is that, if the Software (or any derivative works thereof) is used for any of your commercial products or services that have more than 100 million monthly active users, or more than 20 million US dollars (or equivalent in other currencies) in monthly revenue, you shall prominently display &quot;Kimi K2.5&quot; on the user interface of such product or service.
        • ffsm82 hours ago
          My first reaction was &quot;well, who knows how much revenue they&#x27;re actually doing&quot;<p>But at least the rumor mill has them significantly above that line:<p>&gt; Revenue: As of March 2026, reports suggest Cursor has surpassed $2 billion in annualized revenue (ARR).
        • lmc3 hours ago
          That&#x27;s not an open source license, then.
          • kbrkbr2 hours ago
            Why not?
            • lmc1 hour ago
              This &#x27;Modified MIT&#x27; is not a license that has been through the OSI process: <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;The_Open_Source_Definition#Compliant_licenses" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;The_Open_Source_Definition#Com...</a><p>You can&#x27;t just add random terms to an existing license and use its name. &quot;Modified MIT: Like MIT but pay us 50 million dollars.&quot;<p>Perhaps CC-BY would&#x27;ve been more appropriate.
              • igravious1 hour ago
                Correct again -- CC- applies to data, not code -- weights are data, open weights suggests a creative commons approach …<p>“ CC-BY 4.0 Creative Commons Attribution 4.0 International<p>This license requires that reusers give credit to the creator. It allows reusers to distribute, remix, adapt, and build upon the material in any medium or format, even for commercial purposes.<p>BY Credit must be given to you, the creator. ”<p>it&#x27;s annoying the open source term is being cargo-culted around and I hate to say it but that ship looks like it has sailed.<p>funny that free software people were infuriated by the open source term and now the open source term is being completely misused in another context
          • bakugo3 hours ago
            It wouldn&#x27;t be regardless, because the model is open weights, not open source. It&#x27;s just a license.
            • lmc2 hours ago
              Which contradicts what they say on their website.
              • igravious1 hour ago
                Correct. (and I know you already know this but just for the record: (Nearly?) Everybody abuses the term &quot;open source&quot; when it comes to models. OSI have a post about it: <a href="https:&#x2F;&#x2F;opensource.org&#x2F;ai&#x2F;open-weights" rel="nofollow">https:&#x2F;&#x2F;opensource.org&#x2F;ai&#x2F;open-weights</a>
    • rfoo3 hours ago
      TBH they really shouldn&#x27;t have posted such a tweet in the first place, just sit back and watch their license enforced by the Internet.<p>I had the question &quot;how do you even enforce this weird license term&quot; back then, I guess I know the answer now.
  • gillesjacobs3 hours ago
    Cursor is mostly an IDE &#x2F; coding-agent harness company. So it probably makes sense for them not to train their own base model, but instead license something like Kimi and fine-tune it for their own harness and workflows.<p>Their moat looks pretty thin. A VSCode fork with an open-source LLM fork on top. In the fast-moving coding-agent market, it’s not obvious they keep their massive valuation forever.
    • jstummbillig3 hours ago
      There is a plausible scenario in which software engineering requires a very finite amount of intelligence, in which sota models will be used mainly for other things and where for coding the harness will become increasingly more important than the model.
      • merlindru41 minutes ago
        i&#x27;ve kinda had this thought before but never could express it (&quot;you only need up to a certain level of smartness to express most coding concepts correctly&quot;)<p>but it never occurred to me that, if true, of course the harness becomes increasingly more important. which feels absolutely correct of course.<p>not sure if the hypothesis is even true though.
    • NitpickLawyer3 hours ago
      &gt; Their moat looks pretty thin.<p>Their value is in the data they&#x27;ve collected and are collecting. Usage, acceptance rate, and all the connected signals. Plus having a large userbase where they can A &#x2F; B test any finetune they create.
      • CharlieDigital3 hours ago
        That&#x27;s every harness including VC Code Copilot.<p>People home about Teams sucking, but its market share is several times that of Slack because of distribution.<p>I guarantee that Microsoft has even more data.
      • _puk3 hours ago
        There were conversations in the team yesterday about how Cursor&#x27;s cloud agents are still ahead of Claude from a UX perspective.<p>Obviously we&#x27;re running both, using the right tool for the job.<p>There is stickiness there from being early. That will be hard to replicate.
  • 827a51 minutes ago
    This is exactly what Cursor should be doing, within the obvious bounds of the law and such. Not everyone needs a pristine foundation model. What a waste of compute. Anthropic &amp; OpenAI need product-level competition to knock them off their $25&#x2F;Mtok horse.
  • prodigycorp3 hours ago
    There are many reasons to make fun of Cursor. <i>However</i>, one of the things get right is their autocomplete model.<p>Are there any open models that come close? Why doesnt OAI or Anthropic dedicate some resources to blowing Cursor&#x27;s model out of the water? Cursor&#x27;s completion model is a sticking point for a lot of users.
    • druskacik39 minutes ago
      I agree, their autocomplete (tab) model is the best, but recently I realised I am using it less and less - the new models are so good that I mostly just do agentic coding, and I do very little changes in the codebase by myself. This is probably a general trend and if the usage of autocomplete models is dying out, it&#x27;s understandable the companies are not investing resources into it.
    • seunosewa2 hours ago
      Antigravity has an autocomplete model too. Based on Windsurf&#x27;s, I guess.
      • harmonic1837448 minutes ago
        Absolutely not. Windsurf also just stole an open source model, there’s almost zero chance Google is using that under the hood.
    • olejorgenb2 hours ago
      The model is great. The UX is ~~horrible~~ annoying...
    • g947o2 hours ago
      Most companies don&#x27;t do auto competition these days, including some that just recently stopped offering completion.<p>Which I find very unfortunate. There are so many cases, especially in proprietary codebases with non standard infrastructure, where good autocomplete is much better than &quot;agentic&quot; edits that produce nothing but slop which takes longer to clean up.
  • coreyburnsdev3 minutes ago
    is this the model used on free mode?
  • HeavyStorm2 hours ago
    There&#x27;s no &quot;just&quot; in RL. Fine tuning is very important and could make a lot of difference.
    • merlindru37 minutes ago
      apparently GPT-5 uses the same pretrain as 4o did, hah
  • justindotdev3 hours ago
    im pretty sure this is in violation of moonshot&#x27;s ToS. this is going to be fun to watch unfold
    • NitpickLawyer3 hours ago
      There is no ToS at play here. There&#x27;s only the license[1], which is MIT modified like so:<p>&gt; Our only modification part is that, if the Software (or any derivative works thereof) is used for any of your commercial products or services that have more than 100 million monthly active users, or more than 20 million US dollars (or equivalent in other currencies) in monthly revenue, you shall prominently display &quot;Kimi K2.5&quot; on the user interface of such product or service.<p>[1] - <a href="https:&#x2F;&#x2F;huggingface.co&#x2F;moonshotai&#x2F;Kimi-K2.5&#x2F;blob&#x2F;main&#x2F;LICENSE" rel="nofollow">https:&#x2F;&#x2F;huggingface.co&#x2F;moonshotai&#x2F;Kimi-K2.5&#x2F;blob&#x2F;main&#x2F;LICENS...</a>
      • zozbot2343 hours ago
        Yes, this is pretty clear-cut. There&#x27;s even a great alternative, namely GLM-5, that does <i>not</i> have such a clause (and other alternatives besides) so it feels a bit problematic that they would use Kimi 2.5 and then disregard that advertisement clause.
        • NitpickLawyer3 hours ago
          I&#x27;ve replied down the thread, but there are ways to go around that clause entirely, even if it would be enforceable. The obvious way is to have another company do the modification.
          • zozbot2343 hours ago
            The worthwhile question AIUI is whether AI weights are even protected by human copyright. Note that firms whose &quot;core&quot; value is their proprietary AI weights don&#x27;t even need this (at least AIUI) since they always can fall back on &quot;they are clearly protected against misappropriation, like a trade secret&quot;. It becomes more interesting wrt. openly available AI models.
            • Majromax52 minutes ago
              &gt; The worthwhile question AIUI is whether AI weights are even protected by human copyright.<p>I&#x27;m also deeply curious about this legal question.<p>As I see it, model weights are the result of a mechanistic and lossy translation between training data and the final output weights. There is some human creativity involved, but that creativity is found exclusively in the model&#x27;s code and training data, which are independently covered by copyright. Training is like a very expensive compilation process, and we have long-established that compiled artifacts are not distinct acts of creation.<p>In the case of a proprietary model like Kimi, copyright might survive based on &#x27;special sauce&#x27; training like reinforcement learning – although that competes against the argument that pretraining on copyrighted data is &#x27;fair use&#x27; transformation. However, I can&#x27;t see a good argument that a model trained on a fully public domain dataset (with a genuinely open-source architecture) could support a copyright claim.
    • kgeist2 hours ago
      At the same time, Moonshot violated Anthropic&#x27;s ToS by training on their models&#x27; outputs :) [0]. And Anthropic violated copyright law by training on copyrighted material. It&#x27;s violations all the way down.<p><a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=47126614">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=47126614</a>
      • _aavaa_2 hours ago
        I thought Anthopoc’s training was deemed fair used. It was the downloading that was illegal
    • gillesjacobs3 hours ago
      They probably licensed it. Still a bit deceptive not to mention it on the model card&#x2F;blog post, but companies whitelabel all the time without mentioning.<p>It goes against the ML community ethos to obscure it, but is common branding practice.
      • deaux3 hours ago
        No they didn&#x27;t [0][1]. With this leak they&#x27;re probably negotiating as we speak, which could be why they&#x27;ve deleted the posts.<p>[0] <a href="https:&#x2F;&#x2F;chainthink.cn&#x2F;zh-CN&#x2F;news&#x2F;113784276696010804" rel="nofollow">https:&#x2F;&#x2F;chainthink.cn&#x2F;zh-CN&#x2F;news&#x2F;113784276696010804</a><p>[1] <a href="https:&#x2F;&#x2F;pbs.twimg.com&#x2F;media&#x2F;HD2Ky9jW4AAAe0Y?format=jpg&amp;name=medium" rel="nofollow">https:&#x2F;&#x2F;pbs.twimg.com&#x2F;media&#x2F;HD2Ky9jW4AAAe0Y?format=jpg&amp;name=...</a>
        • gillesjacobs18 minutes ago
          I stand corrected, that is pretty scummy.<p>I bet Moonshot is going to make them open their wallets to avoid legal trouble.
    • charcircuit3 hours ago
      Kimi K2.5 was released under a modified MIT license (100M+ MAU or $20M+ MRR has to prominently display Kimi K2.5). It will be fine.
      • antirez3 hours ago
        Basically this is true for most startups in the world BUT Cursor, so here you are kinda inverting the logic of the matter. Cursor is at a size that, if they wanted to use K2.5, they could clearly state that it was K2.5 or get a license to avoid saying it.
        • NitpickLawyer3 hours ago
          <i>IF</i> we assume that the modified MIT clause is enforceable. And <i>if</i> we assume Cursor Inc. is running the modification. It could very well be the case that Cursor Research LTD is doing the modifications and re-licensing it to Cursor Inc. That would make any clause in the modified MIT moot.
  • simonw1 hour ago
    I&#x27;m annoyed that we still don&#x27;t know for certain which base model they used for Cursor 1.<p>This feels really rude to me. I have no problem with them fine-tuning open weight models to create their own - they are getting great results, and Cursor&#x27;s research term should be respected for that. But deliberately hiding the base model they use is disrespectful of the researchers who created <i>that</i> model.
  • olejorgenb3 hours ago
    To be fair, is &quot;with RL&quot;, &quot;just&quot;?<p>They should have disclosed it though. If they didn&#x27;t it&#x27;s a bad look for sure.
  • todteera3 hours ago
    From a users perspective, do we really care what model we&#x27;re using under the hood? Or how well the software is solving our problems?<p>Seems like cursor is trying to build a &quot;thicker wrapper&quot; beyond the harness. Either to protect against Anthropic shutting them off or increase margins.
  • vachina3 hours ago
    A question. I’m due for a yearly Cursor subscription renewal, how does the credit limit look like?<p>Currently I’ve not hit any of the limits despite using it quite rigorously, I wonder if this will change with a renewal?
    • thewhitetulip1 hour ago
      But you have to buy into it right? If you don&#x27;t have a limit then what did your contract look like?
  • taytus1 hour ago
    YC is back at it again.
  • chvid2 hours ago
    Moonshot is raising money at a 10B usd valuation, cursor&#x2F;anysphere is at a 30B usd valuation.
  • Sammi4 hours ago
    As a paying customer, it just doesn&#x27;t feel good that they are trying to pass off someone else&#x27;s model as their own.<p>I mean I guess this is what businesses do all the time. There&#x27;s a term for it even, it&#x27;s called white-labeling.<p>But is this all that Cursor have? They pass of VS Code as their own, they pass off Kimi as their own... What do Cursor even do? What do I need them for?
    • jstanley3 hours ago
      As a paying customer, I don&#x27;t care where the model comes from, I only care how good it is.
      • Sammi3 hours ago
        Sure, and also at what price point.<p>But can I rely on Cursor to be able to keep delivering, when they aren&#x27;t the one&#x27;s doing the work themselves?
    • khuey3 hours ago
      White-labeling may be slightly dishonest to the consumer but the manufacturer and distributor are honest with each other. That doesn&#x27;t appear to be the case here (Kimi&#x27;s license requires publicly acknowledging Kimi is used for anyone operating at Cursor&#x27;s scale).
  • mono4423 hours ago
    This whole ai stuff feels like a big bubble especially with the oil price soon at $200 and guaranteed recession.
  • cbg02 hours ago
    Scores higher than Opus 4.6 on their in-house benchmark? Sounds legit.
  • EugeneOZ1 hour ago
    I don&#x27;t know - it works okay (yet to be tested whether it is actually smarter than Opus 4.6), but it is not bad at all. So far, it works quite fine (I&#x27;m not testing the &quot;fast&quot; version).
  • lossolo3 hours ago
    Their first model was also based on an open source Chinese base model. They never fully trained their own model.
  • DeathArrow3 hours ago
    I whish it was GLM 5.0.
  • koakuma-chan3 hours ago
    Cursor can&#x27;t compete with Claude Code&#x27;s subsidized pricing, so they are trying to gaslight people that their cheap model is good enough.
  • rvz3 hours ago
    Honestly I don&#x27;t think this leak is any good for Cursor. Not only this appears as a violation to Moonshot&#x27;s ToS, this may also be in fact enough evidence for Anthropic to ban Cursor from using their models, just like they are doing to OpenCode.<p>Why? As I said before, Anthropic mentions Moonshot AI (Maker of the Kimi models) as one of the AI labs that were part of this alleged &quot;distillation attack&quot; [0] campaign and will use that reason to cut off Cursor, Just like they did to OpenAI, xAI and OpenCode.<p>Let&#x27;s see if the market thinks Composor 2 is <i>really</i> that good without the Claude models helping Cursor. (If Anthropic cuts them off).<p>[0] <a href="https:&#x2F;&#x2F;www.anthropic.com&#x2F;news&#x2F;detecting-and-preventing-distillation-attacks" rel="nofollow">https:&#x2F;&#x2F;www.anthropic.com&#x2F;news&#x2F;detecting-and-preventing-dist...</a>
    • Majromax47 minutes ago
      &gt; this may also be in fact enough evidence for Anthropic to ban Cursor from using their models, just like they are doing to OpenCode.<p>The Anthropic ban on OpenCode isn&#x27;t an <i>Anthropic</i> ban on OpenCode, it&#x27;s a ban on using a Calude Code subscription with OpenCode. That&#x27;s justified (or not) under various ToS arguments, but one can still use OpenCode with the more expensive API access.<p>Anthropic&#x27;s complaint about distillation attacks is a distinct prong, one not levied against OpenCode. Additionally, the distillation activities described in your link don&#x27;t describe Cursor&#x27;s routine use of Anthropic&#x27;s models. There, the model outputs are a primary product (e.g. the autocompleted code), and any learning signals provided are incidental.
      • zozbot23442 minutes ago
        Anthropic&#x27;s complaint about &quot;distillation&quot; attacks (obligatory scare quotes because training on glorified chat logs is a far cry from actually distilling from model weights you have real access to) is also about ToS violations. Anthropic&#x27;s ToS, like OpenAI&#x27;s, forbids you from exploiting interactions with their model for the purpose of building a competitor, even though rumor has it that the AI industry has been doing exactly this for a long time anyway.
    • charcircuit3 hours ago
      Kimi K2.5 is an open source model. It is intended for people to make derivative models.
  • heliumtera2 hours ago
    For all the muh productivity guys that like to claim they can turn invisible when no one is looking, an produce 600k lock over 6 weeks, well...cursor is useless now. We know kimi K2.5 won&#x27;t make you 100 trillion times faster.<p>Cursor is killed for this market.
  • agluszak3 hours ago
    A hyped startup providing zero added value, burning investor money only to repackage somebody else&#x27;s work? That&#x27;s new... &#x2F;s
    • DeathArrow2 hours ago
      It depends on what you consider value. People are using it so they find some value.