Meta Muse Glimmer – open weights 30B local coding model

(research.meta.ai)

351 points | by riordan 3 hours ago

26 comments

  • avaer 1 hour ago
    I lament the comments saying this in any way redeems Meta (the company).

    The researchers releasing this stuff have almost nothing to do with Meta other than being bankrolled by the slaughterhouse.

    You aren't the customer, you are the pawn in big tech's game of thrones. Your good will is a commodity to be traded, almost literally. It will be used against you the moment it's convenient. This is open weights because Meta couldn't monetize it in any other way than to cloud developer's judgement of their reputation.

    But I guess most people just don't care.

    I'm glad it's open. It does not make me think any better of Meta.

    • jjice 6 minutes ago
      I'd also argue this is the case for any company releasing open weights. They're not righteous, they're marketing. That's not necessarily a bad thing! They're releasing some great stuff for free and we benefit from that. Every company doing this has a motivation to not release these for free.

      Alibaba, Google, Moonshot, Thinking Machines, etc are not releasing their models for free because they love to. They want to grab market share. I'll take it.

      I still will not use a hosted Meta product, but damn this model looks solid.

    • monster_truck 36 minutes ago
      Meta can never be redeemed, but it's still valid to admit that FB at one point had a very badass engineering culture.

      They're one of 2 companies I would absolutely never work for (weapons etc aside). FB's recruiters hounded me so often I requested that they blackball me. The day they became Meta, I learned this by checking my email to see that they started trying to reach out again. I once again requested that they blackball me. This by extention taints OAI, the other company I'll never work for.

      After a few hours with Glimmer I'm pretty impressed. It's better than the benchmark scores seem to indicate compared to Qwen 3.6 27B. I'm very excited for 3.8

      • swiftcoder 22 minutes ago
        > FB at one point had a very badass engineering culture

        Perpetually kneecapped by one of the worst management cultures I've ever seen

      • LorenDB 11 minutes ago
        What is the other company that you would never work for?
    • commoner 49 minutes ago
      Muse Glimmer doesn't redeem Meta, but it's a contribution to the commons and the Apache 2.0 licensing is an improvement from the restricted licenses attached to Llama. If even Meta can use a permissive license for its model weights, so can any other company.
    • mirekrusin 39 minutes ago
      There is literally not a single comment like this, the only off topic comment like this is yours.
    • armchairhacker 14 minutes ago
      You can say the same about planet Earth.
    • captainbland 1 hour ago
      To be honest the main issue with meta has never been around open/closed software. They've also done react, Cassandra and some other bits. But this, like their open weights is like a feather pressing down on the scale compared to things like promoting genocide in Myanmar, enabling Cambridge analytica, creating a huge closed ecosystem which dominate(s/d) local community communication, mandating doxxed communication, trying to replace actual community communication with algorithmic nonsense etc.
    • skinfaxi 35 minutes ago
      How is this non-sequitor the top comment?
      • blackoil 8 minutes ago
        Certain topics bring out the hidden Reddit inside.
      • simianwords 16 minutes ago
        virtue signalling?
        • mrloopex 14 minutes ago
          Think real hard about that. What does it mean if the only hacker chat group on the planet despises meta this much? Think.
    • hn_submit 16 minutes ago
      I can't take any Big Tech company that still uses PHP seriously. Sorry.
    • larodi 54 minutes ago
      Meta and its products, as a whole, is a threat to your kids, your mental health, your community's health and the planet as a whole. It is just sad and very repulsive everyone fell so easily addicted to their social drug. Yes - it is a drug, and it is hard to get off from.

      Nothing redeems them at this point of time, they are doing exactly ZERO to redeem. Tossing open weight models (not opensource!!) is not a basis for redemption, and does not constitute remorse in any way. Trying to portray it as such is complicity to META's crimes against humanity.

      • foobar_______ 4 minutes ago
        Social media, often owned and perpetuated by Meta, has poisoned the world. It is not redeemable at this point.
      • Grombobulous 34 minutes ago
        I think it’s also worth pointing out that that there are numerous less evil options to choose from.

        Perhaps none of the AI companies are shining examples of high ethics, but basically all of them have ethical high ground over Meta.

        At least Anthropic isn’t sending private videos from pervert glasses to contract workers in Africa. It’s a low bar but it’s a bar nonetheless.

        • monster_truck 18 minutes ago
          They're still using Elon's servers, though. Not like their models are any good anyways
        • dannyw 15 minutes ago
          I personally don’t like, and wouldn’t work for Meta; but it’s an Apache 2.0 model.

          I’m liking it, and I don’t see a personal moral contradiction here. Do you use React for frontend for example?

          I also wish this HN post is a bit more focused on the release, and less noise around Meta.

    • keybored 37 minutes ago
      Any retort to do this like “but why would they just openly release this” pretty much answers itself. Public relations.

      If a company can spend money to redeem itself then, well, it can (game theoretically or whatever) do whatever it wants in the future and then spend money to wipe the slate clean.

    • tonyhart7 9 minutes ago
      what makes Meta so bad ???? they just your average billion dollar company
    • darig 1 hour ago
      [dead]
  • scrlk 2 hours ago
    Will be interesting to see how Qwen3.8 27B compares against this once it releases this week. Seems like dense 30B is back in fashion?

    EDIT: An open weight version of Muse Spark 1.2 is going to be released as well:

    https://x.com/alexandr_wang/status/2086756152034066792

    https://xcancel.com/alexandr_wang/status/2086756152034066792

    • pu_pe 2 hours ago
      Based on the benchmarks, it seems that Muse Glimmer barely edges out against Qwen3.6 27B, except for tool-calling skills (MCP, etc.). I wouldn't be surprised if they released it now because they are afraid they wouldn't beat Qwen3.8 27B.
      • mycall 1 hour ago
        Do AI companies make release plans based on upcoming other models like this? I would think all the processes that go into the repository and weight infrastructure pre-training, checkpointing, knowledge distillation, model compression, post training pipeline, ecosystem integrations, inference API, benchmarking, human eval/safety/alignment, docs, etc... all that dictates the release schedule.
        • pu_pe 12 minutes ago
          Yeah but you can probably have everything ready and then accelerate as necessary. Meta itself did this when releasing Llama 4, it was a really botched release right when they were feeling the heat from DeepSeek and others.
        • michimagdesign 1 hour ago
          Yes, not every model release is reactionary to other labs. Either they had hints for the release of other models or they cut efforts in late stage testing of the models to hit these earlier release dates. There’s always some flexibility. And there’s certainly the incentive to cannibalize the news cycles for competitor models.
          • skohan 19 minutes ago
            I could imagine pulling out all the stops to get a release over the finish line a week early if you're worried about being surpassed by another release
        • echelon 1 hour ago
          There has been a long history of AI model releases made shortly before or after a major planned release by another company. Almost always to upstage or steal thunder.

          Just recently, Minimax H3 released as open weights on the eve of Seedance 2.5 global availability. It's not as good, but it's good enough and it's completely open.

          Flux 3, which is nowhere near as good as either, suddenly announced their release once news of these other two became public. They knew if they waited they'd be ignored. It didn't really help them much, unfortunately.

          The LLM releases are even more rivalrous.

          And don't forget all of the competing launches planned before Google IO or major release events.

          Companies like to eat into the news and press cycle of their rivals.

          • Sabinus 1 hour ago
            I've seen it here on HN (it's particularly noticeable via the /active page) multiple times. If Google, OpenAI or Anthropic release something significant, odds are good you'll see a headline from one of the others.
          • Forgeties79 1 hour ago
            >long history

            Seems a bit premature of a statement lol

            • echelon 1 hour ago
              If you start counting since WaveNet or BERT, it's been ages. Especially when it feels like decades of advancements happen every single year, and rival labs are always trying to one up each other.
              • Forgeties79 29 minutes ago
                I don’t start counting since we WaveNet or BERT so there you go!

                Even if I did, we’re talking barely a decade

        • stogot 1 hour ago
          the last few items there (benchmarking, human evaluation, docs) can be rushed or skipped by leadership if they want to beat comp. they probably spend a few weeks on those things normally
          • dannyw 11 minutes ago
            One window that can be shortened is working with software ecosystem and upstream partners; think day 0 on together, fireworks, Unsloth, etc. That obviously happens from partners getting embargoed weights early.
    • karimf 2 hours ago
      Yes, and also waiting for the next iteration of Gemma. Muse or Qwen are optimized for coding, while IMO Gemma is still better for non-coding tasks.

      https://x.com/osanseviero/status/2086107547535122767

      • dannyw 11 minutes ago
        You can partially tell by the tokeniser; which gives you some hint into the training corpus mix.

        </div> is four Gemma4 tokens, but one Qwen3.6 token.

    • Gecko4072 2 hours ago
      Makes me feel hopeful. Things felt more positive around the llama 3 era. Now it’s like a dark, dreadful race.
    • wronglebowski 2 hours ago
      It’s really interesting timing, Qwen over thinking is what kills it for me. I’m just glad we have more options in this size class now.
      • dannyw 9 minutes ago
        Qwen thinking is really good in Mandarin; and probably natively trained the most there.

        Try a system prompt requiring it to think in Mandarin, while still delivering the response in the user’s language.

    • imilev 1 hour ago
      yes i think everyone is waiting to see that ;d, i've been on qwen for the last year and a half now.
    • ignoramous 1 hour ago
      > Seems like dense 30B is back in fashion?

      Surprising that Meta don't host this model, even as rate-limited free-tier.

      > open weight version of Muse Spark 1.2

      Wait. Is this "version" different from what Meta serves?

    • lostmsu 2 hours ago
      It seems worse than 3.6, but a bit smaller.

      UPD. was wrong on smaller, it's actually much larger

      • IsTom 1 hour ago
        How is 30B smaller than 27B?
        • LeBit 1 hour ago
          It uses fractal compression
        • lostmsu 1 hour ago
          They say it is trained with quantization awareness, so it should only be 15GB or so. Qwen was only trained in FP8 with QAT.

          UPD, NVM, got misled by comments here. It is actually almost 60 GB so much larger

          • ricardobeat 9 minutes ago
            Quantization awareness doesn’t change the size of the weights, just means it won’t degrade when quantized. QAT = quantization aware training. They will both be very similar in size at the same quant.
          • xienze 36 minutes ago
            You're mixing up sizes of different quants. The 60GB is unquantized, and Qwen's unquantized size is around 54GB. Their sizes as like quantization levels are similar.
  • cmiles8 1 hour ago
    With the business model for API based LLMs looking iffy at best it seems like we’re heading back to the “server under your desk” era of IT again.
    • cube00 33 minutes ago
      Considering how all the big players are playing fast [1] and loose [2] with limits, billing [3] and undisclosed changes that burn your tokens on autopilot [4], it can't happen soon enough.

      [1]: Limits may change without notice, including due to capacity constraints. - https://support.google.com/gemini/answer/16275805?sjid=14713....

      [2]: "standard limits" are never defined - https://support.google.com/gemini/answer/16275805?sjid=14713...

      [3]: https://tobyonfitnesstech.com/blog/anthropic-refund-scam/

      [4]: https://news.ycombinator.com/item?id=48947776

      • xscott 20 minutes ago
        Not to mention all the other ways they can screw you:

        - Middle of the day, servers busy? Swap to Sonnet while pretending it's still Opus. Many people won't notice, and nobody can prove anything if they suspect.

        - Middle of the night, server load is light? Put it into extra thinky mode so it burns more tokens to ramp up the bills. Flip the switch where it gets really pedantic about writing lots of extra test cases and verifying against documentation.

        - Demand increases, but don't feel like running more hardware? Switch to low bit quants, but have a monitor model swap back to quality if it can tell you're running a benchmark.

        Assuming model capability plateaus (I think it will), token providers will be in a race to the bottom to maximize profits at the expense of quality that's very difficult to measure.

        • dannyw 7 minutes ago
          These kind of tricks will completely break API customers (although selectively applying it to consumer subs is probably beyond likely at this point).
        • htrp 3 minutes ago
          Ugh... didn't think about extra thinky mode in the middle of the night.

          So many ways for enshittification here.

    • skohan 18 minutes ago
      I've been coding using the LLM server in my living room for the past few weeks, and I haven't had this much fun with tech for ages
  • polymorph1sm 1 hour ago
    Some interesting findings from the chat template designs:

    1. The template name is Onyx ATEM as found in the tool call exception message

    2. It appears to be following a harmony-style chat template. But the tool use seems to be a xml like :<atem:function_calls> / <atem:invoke> / <atem:parameter>

    3. atem: a internal joke of meta in reverse?

    https://huggingface.co/meta-models/Muse-Glimmer-30B/blob/mai...

    • dannyw 1 hour ago
      The XML tags are similar to <antml:xxx>, which is obviously Anthropic ML (or ANTrophic xML).

      I think it’s likely 3; meta in reverse. While tokenisers and preprocessing can catch it, you want your special tokens to be unique and not present in the original corpus. <meta: is likely too common.

  • OsamaJaber 42 minutes ago
    The comparison set is Gemma4-31B and Qwen3.6-27B, not the current Qwen

    Fair on size, but the headline numbers are against a model a generation back

  • _ache_ 2 hours ago
    It is interesting but it does look like a careful distillation of (Spark and) biggers open-weight models.

    The progress compared to Qwen3.6 27B is good, not that impressive, it's a 4 months old model. (kuto to them to compare to 27B dense and not 35B MoE, it's more fair to do so). It is very probable that Qwen3.8 27B will crush Glimmer-30B on most benchmarks.

    • skohan 16 minutes ago
      Still great if they want to play in this space. Having competition for the 24-32GB VRAM target is only good for the end user.
  • sajithdilshan 2 hours ago
    Still needs 32-64GB memory to run it locally. 64GB Macbook pro with an M5 chip costs more than 4k Euros in Germany. A more practical model would be a language specific (e.g Python or JVM language) and excellent at tool calling and reasoning. Maybe that way they can shrink it even more.
    • eigenspace 58 minutes ago
      I think if there's going to be advantages to making smaller, more targeted models, those advantages will probably come from targeting specific domains, not from targeting specific languages.

      I think that if an LLM can't abstract over the differences between Python and C++, it probably will have an even harder time abstracting over the differences between writing code that manages a webserver, and writing code that does aerodynamic simulations.

    • karimf 2 hours ago
      Practically ~20GB with KV cache

      > We quantize weights to ~4-bit, bringing the LM under 20 GB. We validated minimal to no degradation on agentic tasks under compression.

      https://www.reddit.com/r/LocalLLaMA/comments/1vkgsum/introdu...

    • Gecko4072 2 hours ago
      There have been discussions on language specific not really being a relevant change to reduce size.
      • Manfrednotfunny 2 hours ago
        I would love to see any good research projects about it but i have the feeling that Frontier with MoE is making too fast of a progress so that a customized model would always be worse and that the MoE part is actually going somehow in this direction.

        On the other hand, at the GTC was a talk about coding in different lanugage (like spanish) and explaining that the quality between spanish and english is relevant different.

        But i have not found a good article about the impact of learning data with practical experiments or even if the order of the learning data matters.

        At least I think i remember that Meta mentioned having better and less data can be better than more data with lower quality.

        As long as these models can explain to you facts about any other topics, its still overfitted for the task though.

        • mapontosevenths 1 hour ago
          Capability in LLM's is distributed throughout the manifold in subspaces. Even worse, the subspaces exist in superposition.

          That is to say, there is no single 'python' part of the model. The python bit is spread throughout the entire model and overlaps with other pieces that have similar, but unrelated, capabilities. For example the python subpspace might be partially in superposition with cupcake recipes, Esperanto, and calculus. We need calculus in a coding agent but not the other two. However, separating them cleanly is almost impossible, and even identifying them is tough.

          Internally the manifolds are highly inefficient and nothing like you would imagine something humans built would be designed. It's more like something that evolved in nature.

          • Manfrednotfunny 1 hour ago
            My current image from a MoE is that the base/core might be the more generic thing and that things like python are part of one expert though.
            • mapontosevenths 32 minutes ago
              With MOE you train a router designed to select which parts to activate. The router itself is a trained neural network and the 'experts' are usually not really things like 'python'. They're just the functional subspaces I described above.

              Again, those subspaces are all somehow inextricably correlated and live in complex superposition spread throughout the manifold. The router doesn't know (or care) WHY those sections get lit up it just learns which ones to activate to optimize it's own reward function. So maybe it learns to activate "logic", "python" and "cupcake recipes in esperanto" whenever it see's something that kind of looks like python. It's not the best answer, it's just the best answer the tiny router could figure out.

              It's all wildly complicated and inefficient, and works nothing like any reasonable human would imagine that it SHOULD operate.

          • dist-epoch 1 hour ago
            There was some paper about routing at training bio-knowledge into a particular region of the model, which you then can cutoff when serving. But you probably lose some efficiency since maybe you sized that region too small/too big.
            • mapontosevenths 30 minutes ago
              That's a very clever approach. Any idea about the papers title or authors? I'd love to look it up.
    • mihaelm 2 hours ago
      I'm sooo happy I pulled the trigger on upgrading and getting a new laptop (with 64 GB RAM) last summer. Feels like it was just in time before the exponential price jumps.
      • ishtanbul 2 hours ago
        Pulled the trigger?
        • mihaelm 2 hours ago
          lol, you're right, the brainfart completely changes the meaning.

          I corrected it.

        • idiotsecant 2 hours ago
          Common phrase.
          • karolist 2 hours ago
            Parent used "pulled the plug", are you saying it's applicable here and not "pulled the trigger" like suggested?
          • Hinrik 2 hours ago
            That commenter you're replying to knows that. The original commenter before them wrote "pulled the plug" which is different and doesn't quite apply here (actually implies the opposite of what they meant to say).
      • mettamage 2 hours ago
        Bought an M1 64 GB for 2000 euro’s second hand a year ago. That was sweet
        • karolist 2 hours ago
          paid 2.7k € for this same build new in Dec 2023, that was also sweet (still is)
    • dbbk 1 hour ago
      Well if you're spending thousands on API tokens already, you could just drop the same amount on a 128GB MacBook Pro and that's a one time cost.
      • smallerize 1 hour ago
        If you're dropping thousands on API tokens, you're going to be slowed down at least 10x trying to do everything on a single MBP.
        • dannyw 4 minutes ago
          But you could grab a 5090, and paired with some DRAM for MoE offloading of bigger models, and be a happy camper with 1.8TB/s of memory bandwidth.

          Or just use Luna honestly. Worth considering if you’re ok with hosted APIs.

      • Gigachad 25 minutes ago
        The models people are spending thousands on require more on the range of 600-800gb memory.

        128gb hardly runs deepseek v4 flash which is almost free via api pricing.

      • neuroticnews25 1 hour ago
        Don't forget about energy usage, you'll probably never break even vs same model on openrouter.
        • jurgenburgen 1 hour ago
          If you can’t do it cheaper on your own hardware it does make you wonder how much of the cost of inference those large LLM providers are eating? Datacenter hardware isn’t magic.
          • flaunf221 21 minutes ago
            Your personal hardware probably isn't running useful tasks 24/7. If you spend 60% of your 8h work day on full on agentic work, then your hardware is paying off for itself only 20% of available time.
          • petu 55 minutes ago
            Datacenter hardware can batch at large scale, probably over 90% more energy efficient per token than a MacBook.
    • solarkraft 2 hours ago
      I feel like we’ve had this discussion before. From what I remember, specialized models rarely do that much better than general ones, hence no mode Codex models.
    • Archit3ch 57 minutes ago
      > 64GB Macbook pro with an M5 chip costs more than 4k Euros in Germany

      Sure, if you want the latest and almost* greatest. You can pick up an M1 Max 64GB for ~1k.

      * I guess 128GB also exists

    • formerly_proven 1 hour ago
      4K bucks buys you around 180 months of <insert AI subscription here> with zero upfront cost.
      • skohan 14 minutes ago
        If you don't mind exfiltrating all your IP to the API provider
      • zamalek 1 hour ago
        Problem is that might go away or get nerfed.
        • ody4242 1 hour ago
          then you switch provider, it's not a monopoly
      • Mistletoe 1 hour ago
        Haha wow. I’m trying to even imagine the AI landscape in 15 years and I can’t.
    • sparkling 2 hours ago
      Even if you had a 64GB machine: Are you willing to reserve 90% of your memory to run a LLM? With dirt cheap models like deepseek-v4-flash that will run "forever" on $10, the answer for me is clearly: no.
      • Manfrednotfunny 2 hours ago
        I'm waiting for the speed/quality per dollar metric to go down a little bit further and then I will def run it at home.

        Its not just that you send a sentence to an API endpoint, you always send EVERYTHING to that agent as a context.

        You want to analyse your spending history? You now send everything to someone.

        Either no one cares but understands this implication on how easy it is to really capture you or no one really things about it.

        But i'm a lot more diligent on what I send. I disabled the gemini activity feature for example because google started telling me that my stuff could be reviwed by humans.

        • plufz 1 hour ago
          Yeah, it does feel a bit silly with my encrypted disks, encrypted backups, unique passwords, advanced router, etc, while I send everything I do in plain text to anthropic.
      • zoobab 2 hours ago
        "With dirt cheap models like deepseek-v4-flash that will run "forever" on $10, the answer for me is clearly: no."

        When it's free, you are the product.

        • IMTDb 2 hours ago
          Deepseek flash is open weight, this means we can download and run that model without any connection to deepseek, no data/tokens/usage data ever reaches them. They cannot make us their product.
          • Gigachad 23 minutes ago
            All those random api providers are absolutely scooping up your data though. And the hardware to run it locally is absurdly expensive.
        • prplxd_nihilist 2 hours ago
          I see many people saying deepseek and other chinese providers have always been profitable. Also they show their training costs publicly. Can't say for sure since I have not used it personally, but I think they'll for sure outlive the western SOTAs.
          • LogicFailsMe 39 minutes ago
            OpenAI apparently runs a profitable inference business with 40% gross margin, but their advertising budget is nutso and their real costs are pretraining and research. I suspect Deepseek's comp is not predicated on capturing the lightcone of all future value, some googling insinuates their top pay is $212K US which would support that suspicion. Compare and contrast with the $1.35M and up at OpenAI.
        • amrit3128 1 hour ago
          Ah yes, I'm sure Trovalds and Stallman are harvesting my data through free software, aren't they? This argument is used by boomers who were fed cold war era propoganda that surely everybody is selfish, and you're always at fault.
      • halJordan 2 hours ago
        It's the size of a big vm. There's nothing wrong with reserving that much working space for one item.
    • cynicalsecurity 1 hour ago
      I don't understand the desire to run own AI models for programming locally. No laptop is ever going to be as powerful and energy efficient to run anything close to OpenAI, Anthropic or Google models. A model you can run on a loptop is simply not going to work as well as it's needed for programming. Small models for linguistic work fine, but anything more sophisticated simply won't provide enough resources or power. Or models would need to be significantly dumbed down - then why use them at all? So far the idea of carrying a "thin" or "thin"-like device looks more reasonable to me, while running AI on your own server.
      • OtherShrezzing 1 hour ago
        > A model you can run on a loptop is simply not going to work as well as it's needed for programming

        The models you can run on a high-spec laptop today are approximately where frontier models were 12-18mo ago (albeit at a lower tok/s rate). If you scan back through hn comments from that era, you’ll find plenty of people saying “this is powerful enough to massively increase my productivity”.

        • anon373839 34 minutes ago
          > albeit at a lower tok/s rate

          Not always! I get 80-100 tok/s from Qwen 3.6 35B-A3B on a MacBook Pro thanks to MTP. With long contexts that dips to around 50-60. However, prefill is much slower than API models. So it becomes really, really, really critical to not have cache misses.

      • linguae 49 minutes ago
        I’m quite optimistic about the long-term future of local LLMs for privacy and cost control reasons. An LLM running on my own hardware, even if it’s not a laptop but a home server, is one where I don’t need to worry about token limits, token fees, privacy, and “rug-pulling” from the vendor.

        In the short term, the big challenge is being able to afford hardware that can run a ~30B model. Last month I got to experiment with LLMs on a NVIDIA RTX 6000 Ada Generation as a visiting researcher during my summer break. I see the power of local LLMs for agentic coding; they’re no Claude, but they are quite useful. I wish I had gotten into local LLMs before hardware has gotten prohibitively expensive and in some cases unavailable; Apple discontinued certain Mac Minis and Mac Studios with high amounts of RAM due to the RAM shortage.

        Hopefully high RAM prices don’t become a new normal, though the next year or two doesn’t look good.

      • lluisantoni 1 hour ago
        For some companies there might be a need to run them locally. For instance, Apple decided to run LLMs on the phone locally. I guess it depends on how important latency and privacy are. Perhaps Meta is looking at how much interest for those local models is there.
      • ComputerPerson 1 hour ago
        I've never done it but would be interested because it cuts out the burden of worrying about costs. Maybe I'm mistaken on energy cost here. There's a constant raincloud that follows me around regarding limits, and it would be nice to shake that.

        I've been able to accomplish incredible feats (for myself) since GPT-4, so model intelligence is secondary.

      • flaburgan 1 hour ago
        Yet.
  • Gecko4072 2 hours ago
    What I think would be perfect is a model that could run on a single DGX spark and be competitive with DSV4 Flash 731. Flash is already a game changer. Hopefully meta plans on this, like the old 70b. V4 flash is smart enough for any use but slightly too big. 27b-30b isn’t intelligent enough.
    • 127 2 hours ago
      DSV4 Flash 0731 already runs on RTX 4090 24GB + 128GB system RAM at a usable tok/s and quantization.
      • Gecko4072 2 hours ago
        You personally? Just curious. Context window is also a factor and ram isn’t really cheap. Sparks are assembled units which I like.
        • dannyw 1 hour ago
          For the same price as a DGX Spark here (A$8499) I can buy roughly 544GB of DDR5-5200MHz from retail; which on a quad channel platform would deliver ~160gb/s real world; and ~320gb/s with octa channels (Xeon, Threadripper Pro).

          If you can afford it or somehow find a used unit, you can go Epyc for 12 channels.

          8/12 channel DDR5 will beat DGX Spark in inference/decode even without a GPU of any kind, as it’s memory bandwidth bound, and the Spark tops out at ~240gb/s real world.

          With some optimisation and maths, it’s entirely plausible to ach

          You are paying an extraordinary amount of money for the convenience of a super small unit, with still mediocre software support, but at least a community. Expect to be crawling through forum posts regularly, as SM121/Spark has many quirks and ecosystem issues still.

          Please don’t pay another 70-80% gross margins on top of already inflated DRAM prices unless you need. The Spark IS really nice if you want to test out ConnectX or if you really need something small and compact and quiet.

          Also consider: used Adas or even Ampere NVIDIA workstation GPUs can come with a lot of VRAM and be “reasonable”, with CUDA.

    • cmrdporcupine 2 hours ago
      This model I think will be too slow for that on Spark, even at 4 bit quant.

      It's a dense model, not MoE like e.g. Qwen 35b or Gemma 4 26B A4B. On a Spark it will be memory bandwidth limited

      I haven't tried yet (working on it) but back of the napkin estimate puts it at around 15tok/s even after converting to NVFP4. Prefill would be much higher though. That 15tok/sec is pretty typical for dense models of this size:

      NVFP4 Q/K/V/O and MLP projections: ~13 GB/token

      BF16 attention gates: ~3 GB/token

      BF16 LM head: ~2.5 GB/token

      Total: ~18.9 GB/token

      At 273 GB/s, that gives a bandwidth-only ceiling of about 14.5 tok/s; actual performance would be lower.

      • rao-v 1 hour ago
        Native dflash support on day 1 helps a lot! High quality speculative decoding speeds up a lot of agentic work.
  • mirekrusin 16 minutes ago
    Great to see Meta back, looks like really strong, local model, can't wait for llama.cpp support.
  • maxignol 2 hours ago
    Optimizing speed is really the way to go. Yet 24GB is not what everyone can afford. Maybe we could take some of those 56tk/s and transfer into some free RAM space using MoE loading ? I'd be glad with a less than 10GB and more than 6tk/s model.
    • lisplist 2 hours ago
      Unfortunately this is just the entry price for LLMs. With the exception of the Qwen 27B models, I personally haven’t found a ton of use cases for models less than 200B. With the right setup, fine tuning, etc, you can make small models do cool things, but hard to please everyone given the insane hardware costs at the moment and the comparably cheap API costs.
      • dist-epoch 1 hour ago
        Gemma4-E4B (4B params) works pretty well as a local wiki, or when you don't have connectivity.
        • dannyw 1 minute ago
          Nitpick: Gemma4-E4B is actually a 8 billion param model, but only 4.5B params worth of memory bandwidth needed per decode.
    • Manfrednotfunny 2 hours ago
      I don't thinnk just MoE will solve it. If you hit constantly different expert layers, you can't outsource layers efficently and have to swap it in.

      MoE will be faster because it will read less memory for sure, you still have to have it though.

  • richardfey 2 hours ago
    Looking forward to giving this a try with llama.cpp. I’m watching the open-weights competition with high expectations.
  • tosh 3 hours ago
    good to see new open weights releases from meta
    • jauntywundrkind 2 hours ago
      good looking showing too, which is excellent.
    • InfiniteLoup 2 hours ago
      The least they could do, after ruthlessly bombarding my employer's servers with requests, ignoring the robots.txt, scraping everything, and incurring significant Google Maps costs for us in the process.
  • vibe42 1 hour ago
    Meta released their own 4-bit quant of this model for devices with 24GB VRAM.

    That's a modern gaming laptop; cheapest I see in the US with 24GB is $3.5k.

    Should be quite a bit faster than the new M5 MacBook Pro, and you can run Linux on it!

  • eugene3306 16 minutes ago
    will it run on 2x 5060Ti with 16GB each?
    • skohan 12 minutes ago
      It should - the kquant-dynamic variant is targeted towards 32GB. Downloading it now to give it a try.
  • bentt 1 hour ago
    Meta seems like the one American bigtech that would distill the the other American frontier models. My enemy’s enemy is my friend?
    • grim_io 1 hour ago
      They do distill, their own bigger Muse model.
  • gunalx 2 hours ago
    Meta did not abandon opensource. I would love to see a smaller distill, or a moe of this size but the benchmarks seems competetive as long as it isnt benchmaxed witch i would not be suprosed if it is.
    • ignoramous 2 hours ago
      > Meta did not abandon opensource

      Open weights*

      I don't think outside of the Big 3 (Ant, OAI, GDM), given the strong competition from China, any other Lab has a chance at capturing the coding market if they aren't open weights (save for xAI whose latest Grok looks every bit good & will probably rely on Cursor for distribution instead of going open weights). There's literally no other selling point, as the capabilities have mostly converged by now among the chasing pack.

      • ComputerPerson 1 hour ago
        There was a good discussion yesterday on the DeepSeek Flash release thread about this.

        There's a large market, very large, who want the best regardless of what it costs. Probably a large enough market to keep that domain of research afloat (as opposed to shifting research manpower to cost cutting).

        The reasoning is just that the marginal cost of AI is very secondary to fixed costs of the businesses themselves; it's not an excuse to sacrifice performance.

  • Havoc 2 hours ago
    The favourable comparisons to Gemma 4 and qwen3.6 look promising!
    • cmrdporcupine 2 hours ago
      Those two offer MoE variants, this doesn't seem to.

      Dense model makes it dog slow on anything without HBM. Max 15tok/sec on decode on DDR5 systems like a Spark or a Strix Halo -- and that's at 4 bit quant.

      • EddieRingle 1 hour ago
        Dense models run at a very usable speed (Qwen 3.6 was running at ~50t/s last I looked) on my dual 7900 XTX desktop. (And before anyone brings it up, I did not buy them for this purpose, so the up-front cost is irrelevant in my case.)
      • Havoc 39 minutes ago
        The benchmark comparison is against the dense variants not MoE
      • petu 2 hours ago
        3090/4090 probably would do 40 t/s, for 5090 75 t/s is shown in the blog.
  • bronxbomber92 2 hours ago
    I wish they would release the quantized versions in a safetensor format. Many frameworks can't load PTE and GGUF.
  • nutjob2 2 hours ago
    The more open weight models get released the greater the market for personal and small business oriented hardware to run these models. This will drive lower cost hardware, which has stagnated in recent years due to most software not needing the performance and capacity.
    • grim_io 1 hour ago
      Higher demand for 5090's did not make them cheaper, because Nvidia got much higher margin products to focus on.
    • cmrdporcupine 1 hour ago
      The opposite happening because foundries are full to capacity making higher margin stuff.
  • sgt 1 hour ago
    Can I run this on my RTX 5090?
    • skohan 10 minutes ago
      Yes they have quants for 32GB and 20GB use-cases (including mmproj and kv cache + context)
  • solarkraft 2 hours ago
    Wow, Meta is back (at least for now)!

    I like this class of model. Multi-token prediction makes it viable to run dense models at not-too-far-off speeds as MoE models with much better intelligence.

    The submission’s title (open weights 30B local coding model) is luckily wrong: This is meant to be a general agentic model.

    It even comes pre-quantized and with a MTP/drafter model. Looking good!

    Let’s hope they aren’t dishonest with the benchmarks this time …

  • zmmmmm 2 hours ago
    Meta knows how to win back developer's hearts .... let's see if they have the goods
    • xandrius 2 hours ago
      If there is anything meta can do to regain hearts other than owning up their evil deeds, radically change their business model and paying up for taxes and damages, then the world is truly fucked and corporations will continue to win.
  • korykaai 39 minutes ago
    [flagged]
  • jkwang 2 hours ago
    [flagged]
  • moron4hire 25 minutes ago
    "Meta Muse" immediately made me think of Metamucil.

    Product teams really need to hire at least one or two people with a 12-year-old's sense is humor. They need to winnow all the potential stupid jokes out of their product namings.

  • petcat 2 hours ago
    As an industry, I wish we would stop calling these things "open weight" because it is too easy to confuse with actual "open source", which they are not.

    Photoshop source code+ OSI license = open source

    Photoshop binary you can run on your own computer = open weight

    Photoshop SaaS web app = closed, proprietary (Opus, GPT, etc.)

    "Open weight" models are still just binary blobs that are completely inscrutable. It's like bringing home a dog from the rescue and just hoping that it doesn't have a tendency to bite kids in the face. You just can't know. The only thing that you can do is try to add more training (fine tuning) telling it not to bite kids.

    I don't think the FOSS community has ever accepted this, but somehow we're feeling like it is okay now.

    • microtonal 2 hours ago
      Photoshop source code+ OSI license = open source

      Photoshop binary you can run on your own computer = open weight

      I don't think this is a correct analogy. You are not allowed to distribute modified versions of the Photoshop binary. Most open weight model licenses allow you to make and distribute your own finetunes, etc.

    • craigmart 1 hour ago
      I believe that comparing LLMs with traditional deterministic software is fundamentally misleading. It is extremely difficult to truly interpret what LLMs do internally, and as of now, nobody fully understands it. Even if you trained the LLM yourself, there is no source code you can simply read and learn from.

      Sure, having information about how these models were trained is helpful for reproducibility, but it is basically impossible for anyone without substantial capital and access to the same (likely copyrighted) data to reproduce the model. For normal users, owning the model weights essentially means owning 100% of the model, you can inspect and study the weights in much the same way as the lab that produced the model can, you can modify the weights, and you can use and distribute them if the license allows you to

    • QuadmasterXLII 1 hour ago
      Given an open weights model trained to sometimes bite kids, we can’t train it to not bite kids, even though billions of dollars of research have been thrown at this open problem.

      Given an open weights model trained to never bite kids, you can get it to bite kids with 10 prompts and a linear projection, the known simple algorithm doesn’t even need a backwards pass.

      yay asymmetry!

      • kzrdude 1 hour ago
        Any pointers to more info about that? Sounds interesting.
    • piker 2 hours ago
      It is useful to indicate you can run the weights on your own hardware. That’s categorically different from most other commercial offerings. It’s as if your adobe example ignores the reality that would exist had photoshop been invented in 2019: cloud only.
    • monster_truck 2 hours ago
      This analogy is terrible and seems to be extremely misinformed about how rescues evaluate dogs before they are put up for adoption
      • petcat 2 hours ago
        I am extremely well aware of how rescues evaluate dogs. And I'm also fully aware that they do not know the full history of the dog. They go through a limited set of testing and interrogation to evaluate the safety of the dog. That's it.