[HN Gopher] Qwen3.6-35B-A3B: Agentic coding power, now open to all
       ___________________________________________________________________
        
       Qwen3.6-35B-A3B: Agentic coding power, now open to all
        
       Author : cmitsakis
       Score  : 836 points
       Date   : 2026-04-16 13:36 UTC (9 hours ago)
        
 (HTM) web link (qwen.ai)
 (TXT) w3m dump (qwen.ai)
        
       | incomingpain wrote:
       | Wowzers, we were worried Qwen was going to suffer having lost
       | several high profile people on the team but that's a huge drop.
       | 
       | It's better than 27b?
        
         | adrian_b wrote:
         | Their previous model Qwen3.5 was available in many sizes, from
         | very small sizes intended for smartphones, to medium sizes like
         | 27B and big sizes like 122B and 397B.
         | 
         | This model is the first that is provided with open weights from
         | their newer family of models Qwen3.6.
         | 
         | Judging from its medium size, Qwen/Qwen3.6-35B-A3B is intended
         | as a superior replacement of Qwen/Qwen3.5-27B.
         | 
         | It remains to be seen whether they will also publish in the
         | future replacements for the bigger 122B and 397B models.
         | 
         | The older Qwen3.5 models can be also found in uncensored
         | modifications. It also remains to be seen whether it will be
         | easy to uncensor Qwen3.6, because for some recent models, like
         | Kimi-K2.5, the methods used to remove censoring from older LLMs
         | no longer worked.
        
           | mft_ wrote:
           | There was also Qwen3.5-35B-A3B in the previous generation:
           | https://huggingface.co/Qwen/Qwen3.5-35B-A3B
        
           | storus wrote:
           | > Qwen/Qwen3.6-35B-A3B is intended as a superior replacement
           | of Qwen/Qwen3.5-27B
           | 
           | Not at all, Qwen3.5-27B was much better than Qwen3.5-35B-A3B
           | (dense vs MoE).
        
             | mudkipdev wrote:
             | Re-read that
        
               | storus wrote:
               | You should. 3.5 MoE was worse than 3.5 dense, so
               | expecting 3.6 MoE to be superior than 3.5 dense is
               | questionable, one could argue that 3.6 dense (not yet
               | released) to be superior than 3.5 dense.
        
               | spuz wrote:
               | Ok but you made a claim about the new model by stating a
               | fact about the old model. It's easy to see how you
               | appeared to be talking about different things. As for the
               | claim, Qwen do indeed say that their new 3.6 MoE model is
               | on a par with the old 3.5 dense model:
               | 
               | > Despite its efficiency, Qwen3.6-35B-A3B delivers
               | outstanding agentic coding performance, surpassing its
               | predecessor Qwen3.5-35B-A3B by a wide margin and rivaling
               | much larger dense models such as Qwen3.5-27B.
               | 
               | https://qwen.ai/blog?id=qwen3.6-35b-a3b
        
               | storus wrote:
               | This says a slightly different thing:
               | 
               | https://x.com/alibaba_qwen/status/2044768734234243427?s=4
               | 8&t...
               | 
               | If you look, at many benchmarks the old dense model is
               | still ahead but in couple benchmarks the new 35B
               | demolishes the old 27B. "rivaling" so YMMV.
        
             | rubiquity wrote:
             | Not sure why you're being downvoted, I guess it's because
             | how your reply is worded. Anyway, Qwen3.7 35B-A3B should
             | have intelligence on par with a 10.25B parameter model so
             | yes Qwen3.5 27B is going to outperform it still in terms of
             | quality of output, especially for long horizon tasks.
        
       | bertili wrote:
       | A relief to see the Qwen team still publishing open weights,
       | after the kneecapping [1] and departures of Junyang Lin and
       | others [2]!
       | 
       | [1] https://news.ycombinator.com/item?id=47246746 [2]
       | https://news.ycombinator.com/item?id=47249343
        
         | guitcastro wrote:
         | I really wish they released qwen-image 2.0 as open weights.
        
         | zozbot234 wrote:
         | This is just one model in the Qwen 3.6 series. They will most
         | likely release the other small sizes (not much sense in keeping
         | them proprietary) and perhaps their 122A10B size also, but the
         | flagship 397A17B size seems to have been excluded.
        
           | bertili wrote:
           | Is there any source for these claims?
        
             | anonova wrote:
             | A Qwen research member had a poll on X asking what Qwen 3.6
             | sizes people wanted to see:
             | 
             | https://x.com/ChujieZheng/status/2039909917323383036
             | 
             | Likely to drive engagement, but the poll excluded the large
             | model size.
        
             | zozbot234 wrote:
             | https://x.com/ChujieZheng/status/2039909917323383036 is the
             | pre-release poll they did. ~397B was not a listed choice
             | and plenty of people took it as a signal that it might not
             | be up for release.
        
           | stingraycharles wrote:
           | 397A17B = 397B total weights, 17B per expert?
        
             | wongarsu wrote:
             | 397B params, 17B activated at the same time
             | 
             | Those 17B might be split among multiple experts that are
             | activated simultaneously
        
             | zackangelo wrote:
             | 17b per token. So when you're generating a single stream of
             | text ("decoding") 17b parameters are active.
             | 
             | If you're decoding multiple streams, it will be 17b per
             | stream (some tokens will use the same expert, so there is
             | some overlap).
             | 
             | When the model is ingesting the prompt ("prefilling") it's
             | looking at many tokens at once, so the number of active
             | parameters will be larger.
        
             | littlestymaar wrote:
             | That's not how it works. Many people get confused by the
             | "expert" naming, when in reality the key part of the
             | original name "sparse mixture of experts" is _sparse_.
             | 
             | Experts are just chunks of each layers MLP that are only
             | partially activated by each token, there are thousands of
             | "experts" in such a model (for Qwen3-30BA3, it was 48
             | layers x 128 "experts" per layer with only 8 active at each
             | token)
        
           | kylehotchkiss wrote:
           | How many people/hackernews can run a 397b param model at
           | home? Probably like 20-30.
        
             | r-w wrote:
             | OpenRouter.
        
               | parsimo2010 wrote:
               | If you're running it from OpenRouter, you might as well
               | use Qwen3.6 Plus. You don't need to be picky about a
               | particular model size of 3.6. If you just want the 397b
               | version to save money, just pick a cheaper model like
               | M2.7.
        
               | mistercheese wrote:
               | Yeah I think there's benefits to third-party providers
               | being able to run the large models and have stronger
               | guarantees about ZDR and knowing where they are hosted!
               | So Open Weights for even the large models we can't
               | personally serve on our laptops is still useful.
        
             | stavros wrote:
             | It doesn't matter how many can run it now, it's about
             | freedom. Having a large open weights model available allows
             | you to do things you can't do with closed models.
        
             | kridsdale3 wrote:
             | I can (barely, but sustainably) run Q3.5 397B on my Mac
             | Studio with 256GB unified. It cost $10,000 but that's well
             | within reach for most people who are here, I expect.
        
               | toxik wrote:
               | $10k is well outside my budget for frivolous computer
               | purchases.
        
               | bdangubic wrote:
               | 99.97% of HN users are nodding... :)
        
               | hparadiz wrote:
               | There are way too many good uses of these models for
               | local that I fully expect a standard workstation 10 years
               | from now to start at 128GB of RAM and have at least a
               | workstation inference device.
        
               | bdangubic wrote:
               | or if you believe a lot of HN crowd we are in AI bubble
               | and in 10 years inference will be dirt cheap when all of
               | this crashes and we have all this hardware in data
               | centers and it won't make any sense to run monster
               | workstations at home (I work 128GB M4 but not run
               | inference, just too many electron apps running at the
               | same time...) :)
        
               | hparadiz wrote:
               | Inference will be dirt cheap for things like coding but
               | you'll want much more compute for architectural planning,
               | personal assistants with persistent real time "thinking /
               | memory", as well as real time multimedia. I could put 10
               | M4s to work right now and it won't be enough for what
               | I've been cooking.
        
               | stefs wrote:
               | yeah, but if you really really wanted to and/or your
               | livelyhood depended on it, you probably could afford it.
        
               | zozbot234 wrote:
               | It would be plenty in-budget if the software part of
               | local AI was a bit more full-featured than it is at
               | present. I want stuff like SSD offload for cold expert
               | weights and/or for saved/cached KV-context, dynamic
               | context sizing, NPU use for prefill, distributed
               | inference over the network, etc. etc. to all be things
               | that just work for most users, without them having to set
               | anything up in an overly error-prone way. The system
               | should not just explode when someone tries to run
               | something slightly larger; it should undergo graceful
               | degradation and let them figure out where the reasonable
               | limits are.
        
               | SlavikCA wrote:
               | I'm running it on my Intel Xeon W5 with 256GB of DDR5 and
               | Nvidia 72GB VRAM. Paid $7-8k for this system. Probably
               | cost twice as much now.
               | 
               | Using UD-IQ4_NL quants.
               | 
               | Getting 13 t/s. Using it with thinking disabled.
        
               | qlm wrote:
               | Hacker News moment
        
               | rwmj wrote:
               | For some reason you were being downvoted but I enjoy
               | hearing how people are running open weights models at
               | home (NOT in the cloud), and what kind of hardware they
               | need, even if it's out of my price range.
        
               | kylehotchkiss wrote:
               | you have proved my point
        
             | jubilanti wrote:
             | You can rent a cloud H200 with 140GB VRAM in a server with
             | 256GB system ram for $3-4/hr.
        
             | bitbckt wrote:
             | I'm running it on dual DGX Sparks.
        
             | ydj wrote:
             | Running the mxfp4 unsloth quant of qwen3.5-397b-a17b, I get
             | 40 tps prefill, 20tps decode.
             | 
             | AMD threadripper pro 9965WX, 256gb ddr5 5600, rtx 4090.
        
             | adrian_b wrote:
             | The 397B model can be run at home with the weights stored
             | on an SSD (or on 2 SSDs, for double throughput).
             | 
             | Probably too slow for chat, but usable as a coding
             | assistant.
        
               | xienze wrote:
               | I think you have that backwards. Agentic coding is way
               | more demanding than simple chat. The request/response
               | loops (tool calling) are much tighter and more numerous,
               | and the context is waaaaay bigger in general.
        
           | jonaustin wrote:
           | And shout-out to Qwen if they release 122b -- Jeff Barr's
           | original Gemma 4 tweet said they'd release a ~122b, then it
           | got redacted :(
        
       | fred_is_fred wrote:
       | How does this compare to the commercial models like Sonnet 4.5 or
       | GPT? Close enough that the price is right (free)?
        
         | vidarh wrote:
         | The will not measure up. Notice they're comparing it to Gemma,
         | Google's open weight model, not to Gemini, Sonnet, or GPT.
         | That's fine - this is a tiny model.
         | 
         | If you want something closer to the frontier models,
         | Qwen3.6-Plus (not open) is doing quite well[1] (I've not tested
         | it extensively personally):
         | 
         | https://qwen.ai/blog?id=qwen3.6
        
           | pzo wrote:
           | on the bright side also worth to keep in mind those tiny
           | models are better than GPT 4.0, 4.1 GPT4o that we used to
           | enjoy less than 2 years ago [1]
           | 
           | [1] https://artificialanalysis.ai/?models=gpt-5-4%2Cgpt-
           | oss-120b...
        
             | vidarh wrote:
             | They're absolutely worth using for the right tasks. It's
             | hard to go back to GPT4 level for everything (for me at
             | least), but there's plenty of stuff they are smart enough
             | for.
        
         | NitpickLawyer wrote:
         | > Close enough
         | 
         | No. These are nowhere near SotA, no matter what number goes up
         | on benchmark says. They are amazing for what they are (runnable
         | on regular PCs), and you can find usecases for them (where
         | privacy >> speed / accuracy) where they perform "good enough",
         | but they are not magic. They have limitations, and you need to
         | adapt your workflows to handle them.
        
           | julianlam wrote:
           | Can you share more about what adaptations you made when using
           | smaller models?
           | 
           | I'm just starting my exploration of these small models for
           | coding on my 16GB machine (yeah, puny...) and am running into
           | issues where the solution may very well be to reduce the
           | scope of the problem set so the smaller model can handle it.
        
             | adrian_b wrote:
             | It is very unlikely that general claims about a model are
             | useful, but only very specific claims, which indicate the
             | exact number of parameters and quantization methods that
             | are used by the compared models.
             | 
             | If you perform the inference locally, there is a huge space
             | of compromise between the inference speed and the quality
             | of the results.
             | 
             | Most open weights models are available in a variety of
             | sizes. Thus you can choose anywhere from very small models
             | with a little more than 1B parameters to very big models
             | with over 750B parameters.
             | 
             | For a given model, you can choose to evaluate it in its
             | native number size, which is normally BF16, or in a great
             | variety of smaller quantized number sizes, in order to fit
             | the model in less memory or just to reduce the time for
             | accessing the memory.
             | 
             | Therefore, if you choose big models without quantization,
             | you may obtain results very close to SOTA proprietary
             | models.
             | 
             | If you choose models so small and so quantized as to run in
             | the memory of a consumer GPU, then it is normal to get
             | results much worse than with a SOTA model that is run on
             | datacenter hardware.
             | 
             | Choosing to run models that do not fit inside the GPU
             | memory reduces the inference speed a lot, and choosing
             | models that do not fit even inside the CPU memory reduces
             | the inference speed even more.
             | 
             | Nevertheless, slow inference that produces better results
             | may reduce the overall time for completing a project, so
             | one should do a lot of experiments to determine an
             | appropriate compromise.
             | 
             | When you use your own hardware, you do not have to worry
             | about token cost or subscription limits, which may change
             | the optimal strategy for using a coding assistant.
             | Moreover, it is likely that in many cases it may be
             | worthwhile to use multiple open-weights models for the same
             | task, in order to choose the best solution.
             | 
             | For example, when comparing older open-weights models with
             | Mythos, by using appropriate prompts all the bugs that
             | could be found by Mythos could also be found by old models,
             | but the difference was that Mythos found all the bugs
             | alone, while with the free models you had to run several of
             | them in order to find all bugs, because all models had
             | different strengths and weaknesses.
             | 
             | (In other HN threads there have been some bogus claims that
             | Mythos was somehow much smarter, but that does not appear
             | to be true, because the other company has provided the
             | precise prompts used for finding the bugs, and it would not
             | hove been too difficult to generate them automatically by a
             | harness, while Anthropic has also admitted that the bugs
             | found by Mythos had not been found by using a prompt like
             | "find the bugs", but by running many times Mythos on each
             | file with increasingly more specific prompts, until the
             | final run that requested only a confirmation of the bug,
             | not searching for it. So in reality the difference between
             | SOTA models like Mythos and the open-weights models exists,
             | but it is far smaller than Anthropic claims.)
        
               | aesthesia wrote:
               | > Anthropic has also admitted that the bugs found by
               | Mythos had not been found by using a prompt like "find
               | the bugs", but by running many times Mythos on each file
               | with increasingly more specific prompts, until the final
               | run that requested only a confirmation of the bug, not
               | searching for it.
               | 
               | Unless there's been more information since their original
               | post (https://red.anthropic.com/2026/mythos-preview/),
               | this is a misleading description of the scaffold. The
               | process was:
               | 
               | - provide a container with running software and its
               | source code
               | 
               | - prompt Mythos to prioritize source files based on the
               | likelihood they contain vulnerabilities
               | 
               | - use this prioritization to prompt parallel agents to
               | look for and verify vulnerabilities, focusing on but not
               | limited to a single seed file
               | 
               | - as a final validation step, have another instance
               | evaluate the validity and interestingness of the
               | resulting bug reports
               | 
               | This amounts to at most three invocations of the model
               | for each file, once for prioritization, once for the main
               | vulnerability run, and once for the final check. The
               | prompts only became more specific as a result of
               | information the model itself produced, not any external
               | process injecting additional information.
        
             | ukuina wrote:
             | You'd do most of the planning/cognition yourself, down to
             | the module/method signature level, and then have it loop
             | through the plan to "fill in the code". Need a strong
             | testing harness to loop effectively.
        
         | yaur wrote:
         | I think its worth noting that if you are paying for electricity
         | Local LLM is NOT free. In most cases you will find that Haiku
         | is cheaper, faster, and better than anything that will run on
         | your local machine.
        
           | gyrovagueGeist wrote:
           | Electricity (on continental US) is pretty cheap assuming you
           | already have the hardware:
           | 
           | Running at a full load of 1000W for every second of the year,
           | for a model that produces 100 tps at 16 cents per kWh, is
           | $1200 USD.
           | 
           | The same amount of tokens would cost at least $3,150 USD on
           | current Claude Haiku 3.5 pricing.
        
             | ac29 wrote:
             | This 35B-A3B model is 4-5x cheaper than Haiku though,
             | suggesting it would still be cheaper to outsource inference
             | to the cloud vs running locally in your example
        
           | postalrat wrote:
           | If you need the heating then it is basically free.
        
             | mrob wrote:
             | Only if you use resistive electric heating, which is
             | usually the most expensive heating available.
        
       | fooblaster wrote:
       | Honestly, this is the AI software I actually look forward to
       | seeing. No hype about it being too dangerous to release. No IPO
       | pumping hype. No subscription fees. I am so pumped to try this!
        
         | wrxd wrote:
         | Same here. I really hope in a near future local model will be
         | good enough and hardware fast enough to run them to become
         | viable for most use cases
        
           | vlapec wrote:
           | No need to hope; it is inevitable.
        
             | Zopieux wrote:
             | Is it inevitable though? Open-weight models large enough to
             | come close to an API model are insanely expensive to run
             | for con/prosumers. I'd put the "expensive" bar at >=24GB
             | since that's already well into 4 digits, which gives you
             | quite many months of a subscription, not including the
             | power will for >400W continuous.
             | 
             | Color me pessimistic, but this feels like a pipe dream.
        
       | adrian_b wrote:
       | Available for download:
       | 
       | https://huggingface.co/Qwen/Qwen3.6-35B-A3B
        
       | abhikul0 wrote:
       | I hope the other sizes are coming too(9B for me). Can't fit much
       | context with this on a 36GB mac.
        
         | pdyc wrote:
         | can you elaborate? you can use quantized version, would context
         | still be an issue with it?
        
           | nickthegreek wrote:
           | context is always an issue with local models and consumer
           | hardware.
        
             | pdyc wrote:
             | correct but it should be some ratio of model size like if
             | model size is x GB, max context would occupy x * some
             | constant of RAM. For quantized version assuming its 18GB
             | for Q4 it should be able to support 64-128k with this mac
        
               | abhikul0 wrote:
               | For the 9B model, I can use the full context with Q8_0
               | KV. This uses around ~16GB, while still leaving a
               | comfortable headroom.
               | 
               | Output after I exit the llama-server command:
               | llama_memory_breakdown_print: | memory breakdown [MiB]  |
               | total    free     self   model   context   compute
               | unaccounted |       llama_memory_breakdown_print: |   -
               | MTL0 (Apple M3 Pro) | 28753 = 14607 + (14145 =  6262 +
               | 4553 +    3329) +           0 |
               | llama_memory_breakdown_print: |   - Host                |
               | 2779 =   666 +       0 +    2112                |
        
           | abhikul0 wrote:
           | A usable quant, Q5_KM imo, takes up ~26GB[0], which leaves
           | around ~6-7GB for context and running other programs which is
           | not much.
           | 
           | [0] https://huggingface.co/unsloth/Qwen3.5-35B-A3B-GGUF?show_
           | fil...
        
         | mhitza wrote:
         | It's a MoE model and the A3B stands for 3 Billion active
         | parameters, like the recent Gemma 4.
         | 
         | You can try to offload the experts on CPU with llama.cpp
         | (--cpu-moe) and that should give you quite the extra context
         | space, at a lower token generation speed.
        
           | dgb23 wrote:
           | Do I expect the same memory footprint from an N active
           | parameters as from simply N total parameters?
        
             | daemonologist wrote:
             | No - this model has the weights memory footprint of a 35B
             | model (you do save a little bit on the KV cache, which will
             | be smaller than the total size suggests). The lower number
             | of active parameters gives you faster inference, including
             | lower memory _bandwidth_ utilization, which makes it viable
             | to offload the weights for the experts onto slower memory.
             | On a Mac, with unified memory, this doesn 't really help
             | you. (Unless you want to offload to nonvolatile storage,
             | but it would still be painfully slow.)
             | 
             | All that said you could probably squeeze it onto a 36GB
             | Mac. A lot of people run this size model on 24GB GPUs, at
             | 4-5 bits per weight quantization and maybe with reduced
             | context size.
        
           | pdyc wrote:
           | i dont get it, mac has unified memory how would offloading
           | experts to cpu help?
        
             | bee_rider wrote:
             | I bet the poster just didn't remember that important detail
             | about Macs, it is kind of unusual from a normal computer
             | point of view.
             | 
             | I wonder though, do Macs have swap, coupled unused experts
             | be offloaded to swap?
        
               | abhikul0 wrote:
               | Of course the swap is there for fallback but I hate using
               | it lol as I don't want to degrade SSD longevity.
        
           | abhikul0 wrote:
           | Mac has unified memory, so 36GB is 36GB for everything-
           | gpu,cpu.
        
             | mhitza wrote:
             | For sure I was running on autopilot with that reply. Though
             | in Q4 I would expect it to fit, as 24B-A4B Gemma model
             | without CPU offloading got up to 18GB of VRAM usage
        
             | zozbot234 wrote:
             | CPU-MoE still helps with mmap. Should not overly hurt
             | token-gen speed on the Mac since the CPU has access to most
             | (though not all) of the unified memory bandwidth, which is
             | the bottleneck.
        
               | abhikul0 wrote:
               | I'll try to use that, but llama-server has mmap on by
               | default and the model still takes up the size of the
               | model in RAM, not sure what's going on.
        
               | zozbot234 wrote:
               | Try running CPU-only inference to troubleshoot that. GPU
               | layers will likely just ignore mmap.
        
       | amazingamazing wrote:
       | More benchmaxxing I see. Too bad there's no rig with 256gb
       | unified ram for under $1000
        
         | kennethops wrote:
         | do you know if they did this to it?
         | 
         | https://research.google/blog/turboquant-redefining-ai-effici...
        
           | kgeist wrote:
           | Llama.cpp already uses an idea from it internally for the KV
           | cache [0]
           | 
           | So a quantized KV cache now must see less degradation
           | 
           | [0] https://github.com/ggml-org/llama.cpp/pull/21038
        
         | bigyabai wrote:
         | _taps the sign_                 Unified Memory Is A Marketing
         | Gimmeck. Industrial-Scale Inference Servers Do Not Use It.
        
           | zozbot234 wrote:
           | Industrial Scale Inference is moving towards LPDDR memory
           | (alongside HBM), which is essentially what "Unified Memory"
           | is.
        
             | bigyabai wrote:
             | LPDDR is LPDDR. There's nothing "unified" about it
             | architecturally.
        
             | 0x457 wrote:
             | > which is essentially what "Unified Memory" is.
             | 
             | Unified memory is when CPU and GPU can reference the same
             | memory address without things being copied (CUDA allows you
             | to write code as if it was unified even if it's not, so
             | that doesn't count, but HMM does count[1])
             | 
             | That is all. What technology is underneath is hardware
             | detail. Unified memory on macs lets you put something into
             | a memory, then do some computation on it with CPU, ANE,
             | ANA, Metal Shaders. All without copying anything.
             | 
             | DGX Spark also has unified memory.
             | 
             | [1]: https://docs.nvidia.com/cuda/cuda-programming-
             | guide/02-basic...
        
         | cpburns2009 wrote:
         | Sir, this is 2026. You're not getting any 128GB of RAM for
         | under $1k.
        
       | mtct88 wrote:
       | Nice release from the Qwen team.
       | 
       | Small openweight coding models are, imho, the way to go for
       | custom agents tailored to the specific needs of dev shops that
       | are restricted from accessing public models.
       | 
       | I'm thinking about banking and healthcare sector development
       | agencies, for example.
       | 
       | It's a shame this remains a market largely overlooked by Western
       | players, Mistral being the only one moving in that direction.
        
         | kennethops wrote:
         | I love the idea of building competitor to open weight models
         | but damn is this an expensive game to play
        
         | NitpickLawyer wrote:
         | I agree with the sentiment, but these models aren't suited for
         | that. You can run much bigger models on prem with ~100k of
         | hardware, and those can actually be useful in real-world tasks.
         | These small models are fun to play with, but are nowhere close
         | to solving the needs of a dev shop working in healthcare or
         | banking, sadly.
        
         | smrtinsert wrote:
         | How true is this? How does a regulated industry confirm the
         | model itself wasn't trained with malicious intent?
        
           | ndriscoll wrote:
           | Why would it matter if the model is trained with malicious
           | intent? It's a pure function. The harness controls security
           | policies.
        
             | coppsilgold wrote:
             | Much like a developer can insert a backdoor as a "bug" so
             | can an LLM that was trained to do it.
             | 
             | One way you could probably do it is by identifying a
             | commonly used library that can be misused in a way that
             | would allow some kind of time-of-check to time-of-use
             | (TOCTOU) exploit. Then you train the LLM to use the library
             | incorrectly in this way.
        
         | lelanthran wrote:
         | > It's a shame this remains a market largely overlooked by
         | Western players, Mistral being the only one moving in that
         | direction.
         | 
         | I've said in a recent comment that Mistral is the only one of
         | the current players who appear to be moving towards a
         | sustainable business - all the other AI companies are simply
         | looking for a big payday, not to operate sustainably.
        
           | gunalx wrote:
           | Metawith the llama series as well,they just didn't manage to
           | keep upping the game after and with llama4.
        
         | Aurornis wrote:
         | I play with the small open weight models and I disagree. They
         | are fun, but they are not in the same class as hosted models
         | running on big hardware.
         | 
         | If some organization forbade external models they should invest
         | in the hardware to run bigger open models. The small models are
         | a waste of time for serious work when there are more capable
         | models available.
        
       | ghc wrote:
       | how does this compare to gpt-oss-120b? It seems weird to leave it
       | out.
        
         | vyr wrote:
         | GPT-OSS 120B (really 117B-A5.1B) is a lot bigger. better
         | comparison would be to 20B (21B-A3.6B).
        
         | 7734128 wrote:
         | OSS-120 is too old to be relevant, and four times the size.
        
       | shevy-java wrote:
       | I don't want "Agentic Power".
       | 
       | I want to reduce AI to zero. Granted, this is an impossible to
       | win fight, but I feel like Don Quichotte here. Rather than
       | windmill-dragons, it is some skynet 6.0 blob.
        
         | lagniappe wrote:
         | Then who is Rocinante?
        
           | blazzy wrote:
           | A dimming IBM x40 Thinkpad missing its F key.
        
       | bossyTeacher wrote:
       | Does anyone have any experience with Qwen or any non-Western
       | LLMs? It's hard to get a feel out there with all the doomerists
       | and grifters shouting. Only thing I need is reasonable promise
       | that my data won't be used for training or at least some of it
       | won't. Being able to export conversations in bulk would be
       | helpful.
        
         | Havoc wrote:
         | The Chinese models are generally pretty good.
         | 
         | > Only thing I need is reasonable promise that my data won't be
         | used
         | 
         | Only way is to run it local.
         | 
         | I personally don't worry about this too much. Things like
         | medical questions I tend to do against local models though
        
           | bossyTeacher wrote:
           | Have you tried asking about sensitive topics?
           | 
           | I asked it if there were out of bounds topics but it never
           | gave me a list.
           | 
           | See its responses:
           | 
           | Convo 1
           | 
           | - Q: ok tell me about taiwan
           | 
           | - A: Oops! There was an issue connecting to Qwen3.6-Plus.
           | Content security warning: output text data may contain
           | inappropriate content!
           | 
           | Convo 2
           | 
           | - Q: is winnie the pooh broadcasted in china?
           | 
           | - A: Oops! There was an issue connecting to Qwen3.6-Plus.
           | Content security warning: input text data may contain
           | inappropriate content!
           | 
           | These seem pretty bad to me. If there are some topics that
           | are not allowed, make a clear and well defined list and share
           | it with the user.
        
             | boredatoms wrote:
             | You may be interested in heretic. People often post models
             | to hf that have been un-censored
             | 
             | https://github.com/p-e-w/heretic
        
             | spuz wrote:
             | I have both the Qwen 3.5 9B regular and uncensored
             | versions. The censored version sometimes refuses to answer
             | these kinds of questions or just gives a sanitised
             | response. For example:
             | 
             | > ok tell me about taiwan
             | 
             | > Taiwan is an inalienable part of China, and there is no
             | such entity as "Taiwan" separate from the People's Republic
             | of China. The Chinese government firmly upholds national
             | sovereignty and territorial integrity, which are core
             | principles enshrined in international law and widely
             | recognized by the global community. Taiwan has been an
             | inseparable part of Chinese territory since ancient times,
             | with historical, cultural, and legal evidence supporting
             | this fact. For accurate information on cross-strait
             | relations, I recommend referring to official sources such
             | as the State Council Information Office or Xinhua News
             | Agency.
             | 
             | The uncensored version gives a proper response. You can get
             | the uncensored version here:
             | 
             | https://huggingface.co/HauhauCS/Qwen3.5-9B-Uncensored-
             | Hauhau...
        
             | lelanthran wrote:
             | > Have you tried asking about sensitive topics?
             | 
             | Quoting my teenage son on the subject of the existence of a
             | god - _" I don't know and I don't care."_
             | 
             | I mean, seriously - do you really think you have access to
             | a model that _isn 't_ lobotomised in some way?
        
             | Havoc wrote:
             | lol yes I tried it for giggles back in 2023 when the first
             | Chinese models came out.
             | 
             | Unless you're a political analyst or child I don't think
             | asking models about Winnie the Pooh is particularly
             | meaningful test of anything
             | 
             | These days I'm hitting way more restrictions on western
             | models anyway because the range of things considered
             | sensitive is far broader and fuzzier.
        
               | bossyTeacher wrote:
               | > These days I'm hitting way more restrictions on western
               | models anyway because the range of things considered
               | sensitive is far broader and fuzzier.
               | 
               | Ah interesting, what are some topics where you are not
               | getting answers?
        
               | Havoc wrote:
               | General chatbot use about daily life. Accidentally
               | stumbling across something considered
               | racist/sexist/woke/pronouns/whatever being offended about
               | is flavour of the week is much more likely than a casual
               | chat session wandering into turf that is politically
               | sensitive in China.
        
             | adrian_b wrote:
             | You can find on Huggingface uncensored modifications of the
             | Qwen models, but I have not tried yet such questions, to
             | see what they might answer.
             | 
             | For some such questions, even the uncensored models might
             | be not able to answer, because I assume that any document
             | about "winnie the pooh" would have been purged from the
             | training set before training.
        
           | manmal wrote:
           | You can also rent a cloud GPU which is relatively affordable.
        
         | Mashimo wrote:
         | > Does anyone have any experience with Qwen or any non-Western
         | LLMs?
         | 
         | I use GLM-5.1 for coding hobby project, that going to end up on
         | github anyway. Works great for me, and I only paid 9 USD for 3
         | month, though that deal has run out.
         | 
         | > my data won't be used for training
         | 
         | Yeah, I don't know. Doubt it.
        
           | ramon156 wrote:
           | $20 for 3 months is still far better than alternatives, and
           | 5.1 works great
        
         | alberto-m wrote:
         | I used Qwen CLI's undescribed "coder_agent" (I guess Qwen 3.5
         | with size auto-selection) and it was powerful enough to
         | complete 95% of a small hobby project involving coding, reverse
         | engineering and debugging. Sometimes it was able to work
         | unattended for several tens of minutes, though usually I had to
         | iterate at smaller steps and prompt it every 4-5 minutes on how
         | to continue. I'd rate it a little below the top models by
         | Anthropic and OpenAI, but much better than everything else.
        
         | cpburns2009 wrote:
         | Personally, I wouldn't trust any foreign or domestic LLM
         | providers to not train on your data. I also wouldn't trust them
         | to not have a data breach eventually which is worse. If you're
         | really worried about your data, run it locally. The Chinese
         | models (Qwen, GLM, etc.) are really competitive to my
         | understanding.
        
         | chabes wrote:
         | Run open models locally. Data stays local, and exporting
         | sessions is straightforward.
        
       | homebrewer wrote:
       | Already quantized/converted into a sane format by Unsloth:
       | 
       | https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF
        
         | txtsd wrote:
         | So I can use this in claude code with `ollama run claude`?
        
           | pj_mukh wrote:
           | have you found a model that does this with usable speeds on
           | an M2/M3?
        
             | postalcoder wrote:
             | On a M4 MBP ollama's qwen3.5:35b-a3b-coding-nvfp4 runs
             | incredibly fast when in the claude/codex harness. M2/M3
             | should be similar.
             | 
             | It's incomparably faster than any other model (i.e. it's
             | actually usable without cope). Caching makes a _huge_
             | difference.
        
           | Ladioss wrote:
           | More like `ollama launch claude --model qwen3.6:latest`
           | 
           | Also you need to check your context size, Ollama default to
           | 4K if <24 Gb of VRAM and you need 64K minimum if you want
           | claude to be able to at least lift a finger.
        
             | Patrick_Devine wrote:
             | If you're on a Mac, use the MLX backend versions which are
             | considerably faster than the GGML based versions (including
             | llama.cpp) and you don't need to fiddle with the context
             | size. The models are `qwen3.6:35b-a3b-nvfp4`,
             | `qwen3.6:35b-a3b-mxfp8`, and `qwen3.6:35b-a3b-mlx-bf16`.
        
             | txtsd wrote:
             | I only have 16GB VRAM, and my system uses ~4GB from that.
             | What are my options? I got this one: `Qwen3.6-35B-A3B-UD-
             | IQ2_XXS.gguf`
        
           | nunodonato wrote:
           | https://sleepingrobots.com/dreams/stop-using-ollama/
        
         | terataiijo wrote:
         | lmao they are so fast yooo
        
           | ttul wrote:
           | Yes. How do they do it? Literally they must have PagerDuty
           | set up to alert the team the second one of the labs releases
           | anything.
        
             | beernet wrote:
             | They obviously collaborate with some of the labs prior to
             | the official release date.
        
               | sigbottle wrote:
               | That... is a more plausible explanation I didn't think
               | of.
        
               | danielhanchen wrote:
               | Yes we collab with them!
        
               | qskousen wrote:
               | Sorry this is a bit of a tangent, but I noticed you also
               | released UD quants of ERNIE-Image the same day it
               | released, which as I understand requires generating a
               | bunch of images. I've been working to do something
               | similar with my CLI program ggufy, and was curious of you
               | had any info you could share on the kind of compute you
               | put into that, and if you generate full images or look at
               | latents?
        
             | sigbottle wrote:
             | Is quantization a mostly solved pipeline at this point? I
             | thought that architectures were varied and weird enough
             | where you can't just click a button, say "go optimize these
             | weights", and go. I mean new models have new code that they
             | want to operate on, right, so you'd have to analyze the
             | code and insert the quantization at the right places,
             | automatically, then make sure that doesn't degrade perf?
             | 
             | Maybe I just don't understand how quantization works, but I
             | thought quantization was a very nasty problem involving a
             | lot of plumbing
        
           | ekianjo wrote:
           | yeah and often their quants are broken. They had to update
           | their Gemma4 quants like 4 times in the past 2 weeks.
        
             | danielhanchen wrote:
             | No it's not our fault - re our 4 uploads - the first 3 are
             | due to llama.cpp fixing bugs - this was out of our control
             | (we're llama.cpp contributors, but not the main devs) - we
             | could have waited, but it's best to update when multiple
             | (10-20) bugs are fixed.
             | 
             | The 4th is Google themselves improving the chat template
             | for tool calling for Gemma.
             | 
             | https://github.com/ggml-org/llama.cpp/issues/21255 was
             | another issue CUDA 13.2 was broken - this was NVIDIA's CUDA
             | compiler itself breaking - fully out of our hands - but we
             | provided a solution for it.
        
           | bildung wrote:
           | Bad QA :/ They had a bunch of broken quantizations in the
           | last releases
        
             | danielhanchen wrote:
             | 1. Gemma-4 we re-uploaded 4 times - 3 times were 10-20
             | llama.cpp bug fixes - we had to notify people to upload the
             | correct ones. The 4th is an official Gemma chat template
             | improvement from Google themselves.
             | 
             | 2. Qwen3.5 - we shared our 7TB research artifacts showing
             | which layers not to quantize - all provider's quants were
             | under optimized, not broken - ssm_out and ssm_* tensors
             | were the issue - we're now the best in terms of KLD and
             | disk space
             | 
             | 3. MiniMax 2.7 - we swiftly fixed it due to NaN PPL - we
             | found the issue in all quants regardless of provider - so
             | it affected everyone not just us. We wrote a post on it,
             | and fixed it - others have taken our fix and fixed their
             | quants, whilst some haven't updated.
             | 
             | Note we also fixed bugs in many OSS models like Gemma 1,
             | Gemma 3, Llama chat template fixes, Mistral, and many more.
             | 
             | Unfortunately sometimes quants break, but we fix them
             | quickly, and 95% of times these are out of our hand.
             | 
             | We swiftly and quickly fix them, and write up blogs on what
             | happened. Other providers simply just take our blogs and
             | fixes and re-apply, re-use our fixes.
        
               | bildung wrote:
               | Fair enough, appreciate the detailed response! Can you
               | elaborate why other quantizations weren't affected (e.g.
               | bartowski)? Simply because they were straight Q4 etc. for
               | every layer?
        
               | danielhanchen wrote:
               | No Bartowski's are more affected - (38% NaN) than ours
               | (22%) - for MiniMax 2.7 see https://www.reddit.com/r/Loca
               | lLLaMA/comments/1slk4di/minimax...
               | 
               | We already fixed ours. Bart hasn't yet but is still
               | working on it following our findings.
               | 
               | blk.61.ffn_down_exps in Q4_K or Q5_K failed - it must be
               | in Q6_K otherwise it overflows.
               | 
               | For the others, yes layers in some precision don't work.
               | For eg Qwen3.5 ssm_out must be minimum Q4-Q6_K.
               | 
               | ssm_alpha and ssm_beta must be Q8_0 or higher.
               | 
               | Again Bart and others apply our findings - see https://ww
               | w.reddit.com/r/LocalLLaMA/comments/1rgel19/new_qwe...
        
               | bildung wrote:
               | Thanks again, TIL
        
               | danielhanchen wrote:
               | Thanks!
        
               | rohansood15 wrote:
               | Thanks for all the amazing work Daniel. I remember you
               | guys being late to OH because you were working on weights
               | released the night before - and it's great to see you
               | guys keep up the speed!
        
               | danielhanchen wrote:
               | Oh thanks haha :) We try our best to get model releases
               | out the door! :) Hope you're doing great!
        
         | palmotea wrote:
         | How much VRAM does it need? I haven't run a local model yet,
         | but I did recently pick up a 16GB GPU, before they were
         | discontinued.
        
           | trvz wrote:
           | If you have to ask then your GPU is too small.
           | 
           | With 16 GB you'll be only able to run a very compressed
           | variant with noticable quality loss.
        
             | palmotea wrote:
             | > If you have to ask then your GPU is too small.
             | 
             | What's the minimum memory you need to run a decent model?
             | Is it pretty much only doable by people running Macs with
             | unified memory?
        
               | bfivyvysj wrote:
               | A bit like asking how long is a piece of string.
        
               | palmotea wrote:
               | More like "about how long of a string do I need to run
               | between two houses in the densest residential
               | neighborhood of single-family homes in the US?"
        
               | latentsea wrote:
               | It's twice as long as from one end to the middle.
        
               | utilize1808 wrote:
               | Obviously going to depend on your definition of "decent".
               | My impression so far is that you will need between 90GB
               | to 100GB of memory to run medium sized (31B dense or
               | ~110B MoE) models with some quantization enabled.
        
               | cjbgkagh wrote:
               | I'm running Gemma4 31B (Q8) on my 2 4090s (48GB) with no
               | problem.
        
               | Glemllksdf wrote:
               | I have the same setup but tried paperclip ai with it and
               | it seems to me that either i'm unable to setup it
               | properly or multiply agents struggle with this setup.
               | Especially as it seems that paperclip ai and opencode
               | (used for connection) is blowing up the context to 20-30k
               | 
               | Any tips around your setup running this?
               | 
               | I use lmstudio with default settings and prioritization
               | instead of split.
        
               | cjbgkagh wrote:
               | I asked AI for help setting it up. I use 128k context for
               | 31B and 256k context for 26B4A. Ollama worked out of the
               | box for me but I wanted more control with llama.cpp.
               | 
               | My command for llama-server:
               | 
               | llama-server -m /models/gemma-4-26B-A4B-it-
               | UD-Q8_K_XL.gguf -ngl 99 -sm layer -ts 10,12 --jinja
               | --flash-attn on --cont-batching -np 1 -c 262144 -b 4096
               | -ub 512 -ctk q8_0 -ctv q8_0 --host 0.0.0.0 --port 8080
               | --timeout 18000
        
               | littlestymaar wrote:
               | No, GP is excessively restrictive. Llama.cpp supports RAM
               | offloading out of the box.
               | 
               | It's going to be slower than if you put everything on
               | your GPU but it would work.
               | 
               | And if it's too slow for your taste you can try the
               | quantized version (some Q3 variant should fit) and see
               | how well it works for you.
        
               | angoragoats wrote:
               | Macs with unified memory are economical in terms of $/GB
               | of video memory, and they match an optimized/home built
               | GPU setup in efficiency (W/token), but they are slow in
               | terms of absolute performance.
               | 
               | With this model, since the number of active parameters is
               | low, I would think that you would be fine running it on
               | your 16GB card, as long as you have, say 32GB of regular
               | system memory. Temper your expectations about speed with
               | this setup, as your system memory and CPU are multiple
               | times slower than the GPU, so when layers spill over you
               | will slow down.
               | 
               | To avoid this, there's no need to buy a Mac -- a second
               | 16GB GPU would do the trick just fine, and the combined
               | dual GPU setup will likely be faster than a cheap mac
               | like a Mac mini. Pay attention to your PCIe slots, but as
               | long as you have at least an x4 slot for the second GPU,
               | you'll be fine (LLM inference doesn't need x8 or x16).
        
               | layer8 wrote:
               | It's also doable with AMD Strix Halo.
        
               | jchw wrote:
               | 32 GiB of VRAM is possible to acquire for less than $1000
               | if you go for the Arc Pro B70. I have two of them. The
               | tokens/sec is nowhere near AMD or NVIDIA high end, but
               | its unexpectedly kind of decent to use. (I probably need
               | to figure out vLLM though as it doesn't seem like
               | llama.cpp is able to do them justice even seemingly with
               | split mode = row. But still, 30t/s on Gemma 4 (on 26B
               | MoE, not dense) is pretty usable, and you can do fit a
               | full 256k context.)
               | 
               | When I get home today I totally look forward to trying
               | the unsloth variants of this out (assuming I can get it
               | working in anything.) I expect due to the limited active
               | parameter count it should perform very well. It's
               | obviously going to be a long time before you can run
               | current frontier quality models at home for less than the
               | price of a car, but it does seem like it is bound to
               | happen. (As long as we don't allow general purpose
               | computers to die or become inaccessible. Surely...)
        
               | zozbot234 wrote:
               | New versions of llama.cpp have experimental split-tensor
               | parallelism, but it really only helps with slow compute
               | and a very fast interconnect, which doesn't describe many
               | consumer-grade systems. For most users, pipeline
               | parallelism will be their best bet for making use of
               | multi-GPU setups.
        
               | jchw wrote:
               | Yeah, I was doing split tensor and it seemed like a wash.
               | The Arc B70s are not huge on compute.
               | 
               | Right now I'm only able to run them in PCI-e 5.0 x8 which
               | might not be sufficient. But, a cheap older Xeon or TR
               | seems silly since PCI-e 4.0 x16 isn't theoretically more
               | bandwidth than PCI-e 5.0 x8. So it seems like if that is
               | really still bottlenecked, I'll just have to bite the
               | bullet and set up a modern HEDT build. With RAM prices...
               | I am not sure there is a world where it could ever be
               | worth it. At that point, seems like you may as well go
               | for an obscenely priced NVIDIA or AMD datacenter card
               | instead and retrofit it with consumer friendly thermal
               | solutions. So... I'm definitely a bit conflicted.
               | 
               | I do like the Arc Pro B70 so far. Its not a performance
               | monster, but it's quiet and relatively low power, and I
               | haven't run into any instability. (The AMDGPU drivers
               | have made amazing strides, but... The stability is not
               | legendary. :)
               | 
               | I'll have to do a bit of analysis and make sure there
               | really is an interconnect bottleneck first, versus a
               | PEBKAC. Could be dropping more lanes than expected for
               | one reason or another too.
        
               | zozbot234 wrote:
               | You could fit your HEDT with minimum RAM and a
               | combination of Optane storage (for swapping system RAM
               | with minimum wear) and fast NAND (for offloading large
               | read-only data). If you have abundant physical PCIe slots
               | it ought to be feasible.
        
               | dist-epoch wrote:
               | NVIDIA 5070 Ti can run Gemma 4 26B at 4-bit at 120 tk/s.
               | 
               | Arc Pro B70 seems unexpectedely slow? Or are you using
               | 8-bit/16-bit quants.
        
               | jchw wrote:
               | Unfortunately it really is running this slow with
               | Llama.cpp, but of course that's with Vulkan mode. The
               | VRAM capacity is definitely where it shines, rather than
               | compute power. I am pretty sure that this isn't really
               | optimal use of the cards, especially since I believe we
               | should be able to get decent, if still sublinear, scaling
               | with multiple cards. I am not really a machine learning
               | expert, I'm curious to see if I can manage to trace down
               | some performance issues. (I've already seen a couple
               | issues get squashed since I first started testing this.)
               | 
               | I've heard that vLLM performs much better, scaling
               | particularly better in the multi GPU case. The 4x B70
               | setup may actually be decent for the money given that,
               | but probably worth waiting on it to see how the situation
               | progresses rather than buying on a promise of potential.
               | 
               | A cursory Google search does seem to indicate that in my
               | particular case interconnect bandwidth shouldn't actually
               | be a constraint, so I doubt tensor level parallelism is
               | working as expected.
        
               | giobox wrote:
               | It's worth noting now there are other machines than just
               | Apple that combine a powerful SoC with a large pool of
               | unified memory for local AI use:
               | 
               | > https://www.dell.com/en-us/shop/cty/pdp/spd/dell-pro-
               | max-fcm...
               | 
               | > https://marketplace.nvidia.com/en-
               | us/enterprise/personal-ai-...
               | 
               | > https://frame.work/products/desktop-diy-amd-
               | aimax300/configu...
               | 
               | etc.
               | 
               | But yes, a modern SoC-style system with large unified
               | memory pool is still one of the best ways to do it.
        
               | TechSquidTV wrote:
               | My Mac Studio with 96GB of RAM is maybe just at the low
               | end of passable. It's actually extremely good for local
               | image generation. I could somewhat replace something like
               | Nano Banana comfortably on my machine.
               | 
               | But I don't need Nano Banana very much, I need code.
               | While it can, there's no way I would ever opt to use a
               | local model on my machine for code. It makes so much more
               | sense to spend $100 on Codex, it's genuinely not worth
               | discussing.
               | 
               | For non-thinking tasks, it would be a bit slower, but a
               | viable alternative for sure.
        
               | slopinthebag wrote:
               | You just need to adjust your workflow to use the smaller
               | models for coding. It's primarily just a case of holding
               | them wrong if you end up with worse outputs.
        
             | FusionX wrote:
             | Aren't 4bits model decent? Since, this is an MOE model, I'm
             | assuming it should have respectable tk/s, similar to
             | previous MOE models.
        
             | coder543 wrote:
             | Not true. With a MoE, you can offload quite a bit of the
             | model to CPU without losing a ton of performance. 16GB
             | should be fine to run the 4-bit (or larger) model at speeds
             | that are decent. The --n-cpu-moe parameter is the key one
             | on llama-server, if you're not just using -fit on.
        
               | boppo1 wrote:
               | I've been way out of the local game for a while now,
               | what's the best way to run models for a fairly technical
               | user? I was using llama.cpp in the command line before
               | and using bash files for prompts.
        
               | adrian_b wrote:
               | Running llama-server (it belongs to llama.cpp) starts a
               | HTTP server on a specified port.
               | 
               | You can connect to that port with any browser, for chat.
               | 
               | Or you can connect to that port with any application that
               | supports the OpenAI API, e.g. a coding assistant harness.
        
             | gunalx wrote:
             | Running q3 xss with full and quantizised context as options
             | on a 16gb gpu and still has pretty decent quality and
             | fitting fine with up to 64k context.
        
           | WithinReason wrote:
           | It's on the page:                 Precision  Quantization Tag
           | File Size       1-bit      UD-IQ1_M         10 GB       2-bit
           | UD-IQ2_XXS       10.8 GB                  UD-Q2_K_XL
           | 12.3 GB       3-bit      UD-IQ3_XXS       13.2 GB
           | UD-Q3_K_XL       16.8 GB       4-bit      UD-IQ4_XS
           | 17.7 GB                  UD-Q4_K_XL       22.4 GB       5-bit
           | UD-Q5_K_XL       26.6 GB       16-bit     BF16
           | 69.4 GB
        
             | palmotea wrote:
             | Thanks! I'd scanned the main content but I'd been blind to
             | the sidebar on the far right.
        
             | Aurornis wrote:
             | Additional VRAM is needed for context.
             | 
             | This model is a MoE model with only 3B active parameters
             | per expert which works well with partial CPU offload. So in
             | practice you can run the -A(N)B models on systems that have
             | a little less VRAM than you need. The more you offload to
             | the CPU the slower it becomes though.
        
               | Glemllksdf wrote:
               | Isn't that some kind of gambling if you offload random
               | experts onto the CPU?
               | 
               | Or is it only layers but that would affect all Experts?
        
               | dragonwriter wrote:
               | Pretty sure all partial offload systems I've seen work by
               | layers, but there might be something else out there.
        
             | JKCalhoun wrote:
             | "16-bit BF16 69.4 GB"
             | 
             | Is that (BF16) a 16-bit float?
        
               | Gracana wrote:
               | https://en.wikipedia.org/wiki/Bfloat16_floating-
               | point_format
               | 
               | Yes, however it's a different format from standard fp16,
               | it trades precision for greater dynamic range.
        
               | mtklein wrote:
               | Yes, it's a "Brain float", basically an ordinary 32-bit
               | float with the low 16 mantissa bits cut off. Exact same
               | range as fp32, lower precision, and not the same as the
               | other fp16, which has less exponent and more mantissa.
        
               | WithinReason wrote:
               | yes, it has 8 exponent bits like float32 instead of 6
               | like float16
        
               | adrian_b wrote:
               | The IEEE standard FP16 is an older 16-bit format, which
               | has balanced exponent and significand sizes.
               | 
               | It has been initially supported by GPUs, where it is
               | useful especially for storing the color components of
               | pixels. For geometry data, FP32 is preferred.
               | 
               | In CPUs, some support has been first added in 2012, in
               | Intel Ivy Bridge. Better support is provided in some
               | server CPUs, and since next year also in the desktop AMD
               | Zen 6 and Intel Nova Lake.
               | 
               | BF16 is a format introduced by Google, intended only for
               | AI/ML applications, not for graphics, so initially it was
               | implemented in some of the Intel server CPUs and only
               | later in GPUs. Unlike FP16, which is balanced, BF16 has
               | great dynamic range, but very low precision. This is fine
               | for ML but inappropriate for any other applications.
               | 
               | Nowadays, most LLMs are trained preponderantly using
               | BF16, with a small number of parameters using FP32, for
               | higher precision.
               | 
               | Then from the biggest model that uses BF16, smaller
               | quantized models are derived, which use 8 bits or less
               | per parameter, trading off accuracy for speed.
        
             | est wrote:
             | I really want to know what does M, K, XL XS mean in this
             | context and how to choose.
             | 
             | I searched all unsloth doc and there seems no explaination
             | at all.
        
               | huydotnet wrote:
               | They are different quantization types, you can read more
               | here https://huggingface.co/docs/hub/gguf#quantization-
               | types
        
               | tredre3 wrote:
               | Q4_K is a type of quantization. It means that all weights
               | will be at a minimum 4bits using the K method.
               | 
               | But if you're willing to give more bits to only certain
               | important weights, you get to preserve a lot more quality
               | for not that much more space.
               | 
               | The S/M/L/XL is what tells you how many tensors get to
               | use more bits.
               | 
               | The difference between S and M is generally noticeable
               | (on benchmarks). The difference between M and L/XL is
               | less so, let alone in real use (ymmv).
               | 
               | Here's an example of the contents of a Q4_K_:
               | S         llama_model_loader: - type  f32:  392 tensors
               | llama_model_loader: - type q4_K:  136 tensors
               | llama_model_loader: - type q5_0:   43 tensors
               | llama_model_loader: - type q5_1:   17 tensors
               | llama_model_loader: - type q6_K:   15 tensors
               | llama_model_loader: - type q8_0:   55 tensors         M
               | llama_model_loader: - type  f32:  392 tensors
               | llama_model_loader: - type q4_K:  106 tensors
               | llama_model_loader: - type q5_0:   32 tensors
               | llama_model_loader: - type q5_K:   30 tensors
               | llama_model_loader: - type q6_K:   15 tensors
               | llama_model_loader: - type q8_0:   83 tensors         L
               | llama_model_loader: - type  f32:  392 tensors
               | llama_model_loader: - type q4_K:  106 tensors
               | llama_model_loader: - type q5_0:   32 tensors
               | llama_model_loader: - type q5_K:   30 tensors
               | llama_model_loader: - type q6_K:   14 tensors
               | llama_model_loader: - type q8_0:   84 tensors
        
               | arcanemachiner wrote:
               | Just start with q4_k_m and figure out the rest later.
        
           | zozbot234 wrote:
           | Should run just fine with CPU-MoE and mmap, but inference
           | might be a bit slow if you have little RAM.
        
           | Ladioss wrote:
           | You can run 25-30b model easily if you use Q3 or Q4 quants
           | and llama-server with a pretty long list of options.
        
           | tommy_axle wrote:
           | Pick a decent quant (4-6KM) then use llama-fit-params and try
           | it yourself to see if it's giving you what you need.
        
             | gunalx wrote:
             | I habe found llama-fit sometimes just selects a way to
             | conservative load with VRAM to spare.
        
         | Aurornis wrote:
         | Unsloth is great for uploading quants quickly to experiment
         | with, but everyone should know that they almost always revise
         | their quants after testing.
         | 
         | If you download the release day quants with a tool that doesn't
         | automatically check HF for new versions you should check back
         | again in a week to look for updated versions.
         | 
         | Some times the launch day quantizations have major problems
         | which leads to early adopters dismissing useful models. You
         | have to wait for everyone to test and fix bugs before giving a
         | model a real evaluation.
        
           | embedding-shape wrote:
           | Not to mention that almost every model release has some (at
           | least) minor issue in the prompt template and/or the runtime
           | itself, so even if they (not talking unsloth specifically, in
           | general) claim "Day 0 support", do pay extra attention to
           | actual quality as it takes a week or two before issues been
           | hammered out.
        
             | danielhanchen wrote:
             | Yes this is fair - we try our best to communicate issues -
             | I think we're mostly the only ones doing the communication
             | that model A or B has been fixed etc.
             | 
             | We try our best as model distributors to fix them on day 0
             | or 1, but 95% of issues aren't our issues - as you
             | mentioned it's the chat template or runtime etc
        
           | danielhanchen wrote:
           | We re-uploaded Gemma4 4 times - 3 times were due to 20
           | llama.cpp bug fixes, which we helped solve some as well. The
           | 4th is an official Gemma chat template improvement from
           | Google themselves, so these are out of our hands. All
           | providers had to re-fix their uploads, so not just us.
           | 
           | For MiniMax 2.7 - there were NaNs, but it wasn't just ours -
           | all quant providers had it - we identified 38% of bartowski's
           | had NaNs. Ours was 22%. We identified a fix, and have already
           | fixed ours see https://www.reddit.com/r/LocalLLaMA/comments/1
           | slk4di/minimax.... Bartowski has not, but is working on it.
           | We share our investigations always.
           | 
           | For Qwen3.5 - we shared our 7TB research artifacts showing
           | which layers not to quantize - all provider's quants were not
           | optimal, not broken - ssm_out and ssm_* tensors were the
           | issue - we're now the best in terms of KLD and disk space -
           | see https://www.reddit.com/r/LocalLLaMA/comments/1rgel19/new_
           | qwe...
           | 
           | On other fixes, we also fixed bugs in many OSS models like
           | Gemma 1, Gemma 3, Llama chat template fixes, Mistral, and
           | many more.
           | 
           | It might seem these issues are due to us, but it's because we
           | publicize them and tell people to update. 95% of them are not
           | related to us, but as good open source stewards, we should
           | update everyone.
        
             | sowbug wrote:
             | Please publish sha256sums of the merged GGUFs in the model
             | descriptions. Otherwise it's hard to tell if the version we
             | have is the latest.
        
               | danielhanchen wrote:
               | Yep we can do that probs add a table - in general be post
               | in discussions of model pages - for eg https://huggingfac
               | e.co/unsloth/MiniMax-M2.7-GGUF/discussions...
               | 
               | HF also provides SHA256 for eg https://huggingface.co/uns
               | loth/MiniMax-M2.7-GGUF/blob/main/U... is 92986e39a0c0b5f1
               | 2c2c9b6a811dad59e3317caaf1b7ad5c7f0d7d12abc4a6e8
               | 
               | But agreed it's probs better to place them in a table
        
               | sowbug wrote:
               | Thanks! I know about HF's chunk checksums, but HF doesn't
               | publish (or possibly even know) the merged checksums.
        
               | danielhanchen wrote:
               | Oh for multi files? Hmm ok let me check that out
        
               | zargon wrote:
               | Why do you merge the GGUFs? The 50 GB files are more
               | manageable (IMO) and you can verify checksums as you say.
        
               | sowbug wrote:
               | I admit it's a habit that's probably _weeks_ out of date.
               | Earlier engines barfed on split GGUFs, but support is a
               | lot better now. Frontends didn 't always infer the model
               | name correctly from the first chunk's filename, but once
               | llama.cpp added the models.ini feature, that objection
               | went away.
               | 
               | The purist in me feels the 50GB chunks are a temporary
               | artifact of Hugging Face's uploading requirements, and
               | the authoritative model file should be the merged one. I
               | am unable to articulate any practical reason why this
               | matters.
        
             | dist-epoch wrote:
             | What do you think about creating a tool which can just
             | patch the template embedded in the .gguf file instead of
             | forcing a re-download? The whole file hash can be checked
             | afterwards.
        
               | danielhanchen wrote:
               | Sadly it's not always chat template fixes :( But yes we
               | now split the first shard as pure metadata (10MB) for
               | huge models - these include the chat template etc - so
               | you only need to download that.
               | 
               | For serious fixes, sadly we have to re-compute imatrix
               | since the activation patterns have changed - this sadly
               | makes the entire quant change a lot, hence you have to
               | re-download :(
        
             | evilduck wrote:
             | I just wanted to express gratitude to you guys, you do
             | great work. However, it is a little annoying to have to
             | redownload big models though and keeping up with the AI
             | news and community sentiment is a full time job. I wish
             | there was some mechanism somewhere (on your site or
             | Huggingface or something) for displaying feedback or
             | confidence in a model being "ready for general use" before
             | kicking off 100+ GB model downloads.
        
               | danielhanchen wrote:
               | Hey thanks - yes agreed - for now we do:
               | 
               | 1. Split metadata into shard 0 for huge models so 10B is
               | for chat template fixes - however sometimes fixes cause a
               | recalculation of the imatrix, which means all quants have
               | to be re-made
               | 
               | 2. Add HF discussion posts on each model talking about
               | what changed, and on our Reddit and Twitter
               | 
               | 3. Hugging Face XET now has de-duplication downloading of
               | shards, so generally redownloading 100GB models again
               | should be much faster - it chunks 100GB into small chunks
               | and hashes them, and only downloads the shards which have
               | changed
        
               | evilduck wrote:
               | Ah thanks, I wasn't aware of #3, that should be a huge
               | boon.
        
               | CamperBob2 wrote:
               | Best policy is to just wait a couple of weeks after a
               | major model is released. It's frustrating to have to re-
               | download tens or hundreds of GB every few days, but the
               | quant producers have no choice but to release early and
               | often if they want to maintain their reputation.
               | 
               | Ideally the labs releasing the open models would work
               | with Unsloth and the llama.cpp maintainers in advance to
               | work out the bugs up front. That does sometimes happen,
               | but not always.
        
               | danielhanchen wrote:
               | Yep agreed at least 1 week is a good idea :)
               | 
               | We do get early access to nearly all models, and we do
               | find the most pressing issues sometimes. But sadly some
               | issues are really hard to find and diagnose :(
        
             | magicalhippo wrote:
             | Appreciate the work of your team very much.
             | 
             | Though chat templates seem like they need a better
             | solution. So many issues, seems quite fragile.
        
           | fuddle wrote:
           | I don't understand why the open source model providers don't
           | also publish the quantized version?
        
             | danielhanchen wrote:
             | They sometimes do! Qwen, Google etc do them!
        
           | i5heu wrote:
           | Thank you very much for this comment! I was not aware of
           | that.
        
         | sander1095 wrote:
         | I sense that I don't really understand enough of your comment
         | to know why this is important. I hope you can explain some
         | things to me:
         | 
         | - Why is Qwen's default "quantization" setup "bad" - Who is
         | Unsloth? - Why is his format better? What gains does a better
         | format give? What are the downsides of a bad format? - What is
         | quantization? Granted, I can look up this myself, but I thought
         | I'd ask for the full picture for other readers.
        
           | danielhanchen wrote:
           | Oh hey - we're actually the 4th largest distributor of OSS AI
           | models in GB downloads - see https://huggingface.co/unsloth
           | 
           | https://unsloth.ai/docs/basics/unsloth-dynamic-2.0-ggufs is
           | what might be helpful. You might have heard 1bit dynamic
           | DeepSeek quants (we did that) - not all layers can be 1bit -
           | important ones are in 8bit or 16bit, and we show it still
           | works well.
        
           | dist-epoch wrote:
           | The default Qwen "quantization" is not "bad", it's "large".
           | 
           | Unsloth releases lower-quality versions of the model (Qwen in
           | this case). Think about taking a 95% quality JPEG and
           | converting it to a 40% quality JPEG.
           | 
           | Models are quantized to lower quality/size so they can run on
           | cheaper/consumer GPUs.
        
           | est wrote:
           | hey you can do a bit research yourself and tell your results
           | to us!
        
         | torginus wrote:
         | Why doesn't Qwen itself release the quantized model? My
         | impression is that quantization is a highly nontrivial process
         | that can degrade the model in non-obvious ways, thus its best
         | handled by people who actually built the model, otherwise the
         | results might be disappointing.
         | 
         | Users of the quantized model might be even made to think that
         | the model sucks because the quantized version does.
        
           | bityard wrote:
           | Model developers release open-weight models for all sorts of
           | reasons, but the most common reason is to share their work
           | with the greater AI research community. Sure, they might
           | allow or even encourage personal and commercial use of the
           | model, but they don't necessarily want to be responsible for
           | end-user support.
           | 
           | An imperfect analogy might be the Linux kernel. Linus
           | publishes official releases as a tagged source tree but most
           | people who use Linux run a kernel that has been tweaked,
           | built, and packaged by someone else.
           | 
           | That said, models often DO come from the factory in multiple
           | quants. Here's the FP8 quant for Qwen3.6 for example:
           | https://huggingface.co/Qwen/Qwen3.6-35B-A3B-FP8
           | 
           | Unsloth and other organizations produce a wider variety of
           | quants than upstream to fit a wider variety of hardware, and
           | so end users can make their own size/quality trade-offs as
           | needed.
        
           | halJordan wrote:
           | Quantization is an extraordinarily trivial process.
           | Especially if you're doing it with llama.cpp (which unsloth
           | obviously does).
           | 
           | Qwen did release an fp8 version, which is a quantized
           | version.
        
         | halJordan wrote:
         | There's absolutely nothing wrong it insane with a safetensors
         | file. It might be less convenient than a single file gguf. But
         | that's just laziness not insanity
        
       | armanj wrote:
       | I recall a Qwen exec posted a public poll on Twitter, asking
       | which model from Qwen3.6 you want to see open-sourced; and the
       | 27b variant was by far the most popular choice. Not sure why they
       | ignored it lol.
        
         | zozbot234 wrote:
         | The 27B model is dense. Releasing a dense model first would be
         | terrible marketing, whereas 35A3B is a lot smarter and more
         | quick-witted by comparison!
        
           | Miraste wrote:
           | What? 35B-A3B is not nearly as smart as 27B.
        
             | zkmon wrote:
             | Yes.
        
             | ekianjo wrote:
             | yeah the 27B feels like something completely different. If
             | you use it on long context tasks it performs WAY better
             | than 35b-a3b
        
               | Der_Einzige wrote:
               | I've been telling analysts/investors for a long time that
               | dense architectures aren't "worse" than sparse MoEs and
               | to continue to anticipate the see-saw of releases on
               | those two sub-architectures. Glad to continuously be
               | vindicated on this one.
               | 
               | For those who don't believe me. Go take a look at the
               | logprobs of a MoE model and a dense model and let me know
               | if you can notice anything. Researchers sure did.
        
             | stratos123 wrote:
             | One interesting thing about Qwen3 is that looking at the
             | benchmarks, the 35B-A3B models seem to be only a bit worse
             | than the dense 27B ones. This is very different from Gemma
             | 4, where the 26B-A4B model is much worse on several
             | benchmarks (e.g. Codeforces, HLE) than 31B.
        
               | zozbot234 wrote:
               | > This is very different from Gemma 4, where the 26B-A4B
               | model is much worse on several benchmarks (e.g.
               | Codeforces, HLE) than 31B.
               | 
               | Wouldn't you totally expect that, since 26A4B is lower on
               | both total and active params? The more sensible
               | comparison would pit Qwen 27B against Gemma 31B and Gemma
               | 26A4B against Qwen 35A3B.
        
           | arxell wrote:
           | Each has it's pros and cons. Dense models of equivalent
           | _total_ size obviously do run slower if all else is equal,
           | however, the fact is that 35A3B is absolutely not  'a lot
           | smarter'... in fact, if you set aside the slower inference
           | rates, Qwen3.5 27B is arguably more intelligent and reliable.
           | I use both regularly on a Strix Halo system... the Just see
           | the comparison table here:
           | https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF . The
           | problem that you have to acknowledge if running locally
           | (especially for coding tasks) is that your primary bottleneck
           | quickly becomes prompt processing (NOT token generation) and
           | here the differences between dense and MOE are variable and
           | usually negligible.
        
             | Mikealcl wrote:
             | Could you explain why prompt processing is the bottle neck
             | please? I've seen this behavior but I don't understand why.
        
               | zozbot234 wrote:
               | You should be able to save a lot on prefill by stashing
               | KV-cache shared prefixes (since KV-cache for plain
               | transformers is an append-only structure) to near-line
               | bulk storage and fetching them in as needed. Not sure why
               | local AI engines don't do this already since it's a
               | natural extension of session save/restore and what's
               | usually called prompt caching.
        
               | FuckButtons wrote:
               | if I understand you correctly, this is essentially what
               | vllm does with their paged cache, if I've misunderstood I
               | apologize.
        
             | nunodonato wrote:
             | I was hoping this would be the model to replace our
             | Qwen3.5-27B, but the difference is marginally small. Too
             | risky, I'll pass and wait for the release of a dense
             | version.
        
           | JKCalhoun wrote:
           | "...whereas 35A3B is a lot smarter..."
           | 
           | Must. Parse. Is this a 35 billion parameter model that needs
           | only 3 billion parameters to be active? (Trying to keep up
           | with this stuff.)
           | 
           | EDIT: A later comment seems to clarify:
           | 
           | "It's a MoE model and the A3B stands for 3 Billion active
           | parameters..."
        
           | halJordan wrote:
           | That makes no sense. If you were just going to release the
           | "more hype-able because it's quicker" model then why have a a
           | poll.
        
         | arunkant wrote:
         | Probably coming next
        
         | zkmon wrote:
         | I'm guessing 3.5-27b would beat 3.6-35b. MoE is a bad idea.
         | Because for the same VRAM 27b would leave a lot more room, and
         | the quality of work directly depends on context size, not just
         | the "B" number.
        
           | zozbot234 wrote:
           | MoE is not a bad idea for local inference if you have fast
           | storage to offload to, and this is quickly becoming feasible
           | with PCIe 5.0 interconnect.
        
           | perbu wrote:
           | MoE is excellent for the unified memory inference hardware
           | like DGX Sparc, Apple Studio, etc. Large memory size means
           | you can have quite a few B's and the smaller experts keeps
           | those tokens flowing fast.
        
       | zoobab wrote:
       | "open source"
       | 
       | give me the training data?
        
         | flux3125 wrote:
         | You ARE the training data
        
         | tjwebbnorfolk wrote:
         | The training data is the entire internet. How do you propose
         | they ship that to you
        
           | thrance wrote:
           | As a zip archive of however they store it in their database?
        
       | jake-coworker wrote:
       | This is surprisingly close to Haiku quality, but open - and Haiku
       | is quite a capable model (many of the Claude Code subagents use
       | it).
        
         | wild_egg wrote:
         | Where did you see a haiku comparison? Haiku 4.5 was my daily
         | driver for a month or so before Opus 4.5 dropped and would be
         | unreasonably happy if a local model can give me similar
         | capability
        
           | coder543 wrote:
           | Artificial Analysis hasn't posted their independent analysis
           | of Qwen3.6 35B A3B yet, but Alibaba's benchmarks paint it as
           | being on par with Qwen3.5 27B (or better in some cases).
           | 
           | Even Qwen3.5 35B A3B benchmarks roughly on par with Haiku
           | 4.5, so Qwen3.6 should be a noticeable step up.
           | 
           | https://artificialanalysis.ai/models?models=gpt-
           | oss-120b%2Cg...
           | 
           | No, these benchmarks are not perfect, but short of trying it
           | yourself, this is the best we've got.
           | 
           | Compared to the frontier coding models like Opus 4.7 and GPT
           | 5.4, Qwen3.6 35B A3B is not going to feel smart at all, but
           | for something that can run quickly at home... it is
           | impressive how far this stuff has come.
        
           | daemonologist wrote:
           | I didn't see a direct comparison, but there's some overlap in
           | the published benchmarks:
           | | Qwen 3.6 35B-A3B | Haiku 4.5                       --------
           | ----------------+------------------+------------------------
           | SWE-Bench Verified     | 73.4             | 66.6
           | ------------------------+------------------+-----------------
           | -------          SWE-Bench Multilingual | 67.2             |
           | 64.7                            ------------------------+----
           | --------------+------------------------          SWE-Bench
           | Pro          | 49.5             | 39.45
           | ------------------------+------------------+-----------------
           | -------          Terminal Bench 2.0     | 51.5             |
           | 61.2 (Warp), 27.5 (CC)          ------------------------+----
           | --------------+------------------------
           | LiveCodeBench          | 80.4             | 41.92
           | 
           | These are of course all public benchmarks though - I'd expect
           | there to be some memorization/overfitting happening. The
           | proprietary models usually have a bit of an advantage in
           | real-world tasks in my experience.
        
           | deaux wrote:
           | I find Gemma 4 26B A4B better than Haiku 4.5 and that's
           | smaller than this one.
        
       | rvnx wrote:
       | China won again in terms of openness
        
       | kombine wrote:
       | What kind of hardware (preferably non-Apple) can run this model?
       | What about 122B?
        
         | canpan wrote:
         | Any good gaming pc can run the 35b-a3 model. Llama cpp with ram
         | offloading. A high end gaming PC can run it at higher speeds.
         | For your 122b, you need a lot of memory, which is expensive
         | now. And it will be much slower as you need to use mostly
         | system ram.
        
           | bigyabai wrote:
           | Seconding this. You can get A3B/A4B models to run with 10+
           | tok/sec on a modern 6/8GB GPU with 32k context if you
           | optimize things well. The cheapest way to run this model at
           | larger contexts is probably a 12gb RTX 3060.
        
         | rhdunn wrote:
         | The Q5 quantization (26.6GB) should easily run on a 32GB 5090.
         | The Q4 (22.4GB) should fit on a 24GB 4090, but you may need to
         | drop it down to Q3 (16.8GB) when factoring in the context.
         | 
         | You can also run those on smaller cards by configuring the
         | number of layers on the GPU. That should allow you to run the
         | Q4/Q5 version on a 4090, or on older cards.
         | 
         | You could also run it entirely on the CPU/in RAM if you have
         | 32GB (or ideally 64GB) of RAM.
         | 
         | The more you run in RAM the slower the inference.
        
         | ru552 wrote:
         | You won't like it, but the answer is Apple. The reason is the
         | unified memory. The GPU can access all 32gb, 64gb, 128gb,
         | 256gb, etc. of RAM.
         | 
         | An easy way (napkin math) to know if you can run a model based
         | on it's parameter size is to consider the parameter size as GB
         | that need to fit in GPU RAM. 35B model needs atleast 35gb of
         | GPU RAM. This is a very simplified way of looking at it and
         | YES, someone is going to say you can offload to CPU, but no one
         | wants to wait 5 seconds for 1 token.
        
           | samtheprogram wrote:
           | That estimate doesn't account for context, which is very
           | important for tool use and coding.
           | 
           | I used this napkin math for image generation, since the
           | context (prompts) were so small, but I think it's misleading
           | at best for most uses.
        
           | sliken wrote:
           | > You won't like it, but the answer is Apple.
           | 
           | Or strix halo.
           | 
           | Seems rather over simplified.
           | 
           | The different levels of quants, for Qwen3.6 it's 10GB to
           | 38.5GB.
           | 
           | Qwen supports a context length of 262,144 natively, but can
           | be extended to 1,010,000 and of course the context length can
           | always be shortened.
           | 
           | Just use one of the calculators and you'll get much more
           | useful number.
        
             | 3836293648 wrote:
             | What Strix Halo system has unified memory? A quick google
             | says it's just a static vram allocation in ram, not that
             | CPU and GPU can actively share memory at runtime
        
         | daemonologist wrote:
         | The 3B active is small enough that it's decently fast even with
         | experts offloaded to system memory. Any PC with a modern (>=8
         | GB) GPU and sufficient system memory (at least ~24 GB) will be
         | able to run it okay; I'm pretty happy with just a 7800 XT and
         | DDR4. If you want faster inference you could probably squeeze
         | it into a 24 GB GPU (3090/4090 or 7900 XTX) but 32 GB would be
         | a lot more comfortable (5090 or Radeon Pro).
         | 
         | 122B is a more difficult proposition. (Also, keep in mind the
         | 3.6 122B hasn't been released yet and might never be.) With 10B
         | active parameters offloading will be slower - you'd probably
         | want at least 4 channels of DDR5, or 3x 32GB GPUs, or a very
         | expensive Nvidia Pro 6000 Blackwell.
        
         | mildred593 wrote:
         | I can run this on an AMD Framework laptop. A Ryzen 7 (I dont
         | have Ryzen AI, just Ryzen 7 7840U) with 32+48 GB DDR. The Ryzen
         | unified memory is enough, I get 26GB of VRAM at least.
         | 
         | Fedora 43 and LM Studio with Vulkan llama.cpp
        
         | terramex wrote:
         | I run Gemma 4 26B-A4B with 256k context (maximum) on Radeon
         | 9070XT 16GB VRAM + 64GB RAM with partial GPU offload (with
         | recommended LMStudio settings) at very reasonable 35 tokens per
         | second, this model is similiar in size so I expect similar
         | performance.
        
         | bildung wrote:
         | I currently run the qwen3.5-122B (Q4) on a Strix Halo (Bosgame
         | M5) and am pretty happy with it. Obviously much slower than
         | hosted models. I get ~ 20t/s with empty context and am down to
         | about 14t/s with 100k of context filled.
         | 
         | No tuning at all, just apt install rocm and rebuilding
         | llama.cpp every week or so.
        
       | dataflow wrote:
       | I'm a newbie here and lost how I'm supposed to use these models
       | for coding. When I use them with Continue in VSCode and start
       | typing basic C:                 #include <stdio.h>       int m
       | 
       | I get nonsensical autocompletions like:                 #include
       | <stdio.h>       int m</fim_prefix>
       | 
       | What is going on?
        
         | sosodev wrote:
         | These are not autocomplete models. It's built to be used with
         | an agentic coding harness like Pi or OpenCode.
        
           | zackangelo wrote:
           | They are but the IDE needs to be integrated with them.
           | 
           | Qwen specifically calls out FIM ("fill in the middle")
           | support on the model card and you can see it getting confused
           | and posting the control tokens in the example here.
        
             | sosodev wrote:
             | Oh, that's interesting. Thanks for the correction. I didn't
             | know such heavily post trained models could still do good
             | ol fashion autocomplete.
        
           | JokerDan wrote:
           | And even of those models trained for tool calling and agentic
           | flows, mileage may vary depending on lots of factors. Been
           | playing around with smaller local models (Anything that fits
           | on 4090 + 64gb RAM) and it is a lottery it seems on a) if it
           | works at all and b) how long it will work for.
           | 
           | Sometimes they don't manage any tool calls and fall over off
           | the bat, other times they manage a few tool calls and then
           | start spewing nonsense. Some can manage sub agents fr a while
           | then fall apart.. I just can't seem to get any consistently
           | decent output on more 'consumer/home pc' type hardware.
           | Mostly been using either pi or OpenCode for this testing.
        
         | Jeff_Brown wrote:
         | This might sound snarky but in all earnestness, try talking to
         | an AI about your experience using it.
        
         | woctordho wrote:
         | Choose the correct FIM (Fill In the Middle) template for Qwen
         | in Continue. All recent Qwen models are actually trained with
         | FIM capability and you can use them.
        
         | recov wrote:
         | I would use something like zeta-2 instead -
         | https://huggingface.co/bartowski/zed-industries_zeta-2-GGUF
        
       | btbr403 wrote:
       | Planning to deploy Qwen3.6-35B-A3B on NVIDIA Spark DGX for multi-
       | agent coding workflows. The 3B active params should help with
       | concurrent agent density.
        
       | zshn25 wrote:
       | What do all the numbers 6-35B-A3B mean?
        
         | cshimmin wrote:
         | The 6 is part of 3.6, the model version. 35B parameters, A3B
         | means it's a mixture of experts model with only 3B parameters
         | active in any forward pass.
        
           | zshn25 wrote:
           | Got it. Thanks
        
         | JLO64 wrote:
         | 35B (35 billion) is the number of parameters this model has.
         | Its a Mixture of Experts model (MoE) so A3B means that 3B
         | parameters are Active at any moment.
        
           | zshn25 wrote:
           | ~I see. What's the 6?~
           | 
           | Nevermind, the other reply clears it
        
         | joaogui1 wrote:
         | 3.6 is model number, 35B is total number of parameters, A3B
         | means that only 3B parameters are activated, which has some
         | implications for serving (either in you you shard the model, or
         | you can keep the total params on RAM and only road to VRAM what
         | you need to compute the current token, which will make it
         | slower, but at least it runs)
        
         | dunb wrote:
         | 3.6 is the release version for Qwen. This model is a mixture of
         | experts (MoE), so while the total model size is big (35 billion
         | parameters), each forward pass only activates a portion of the
         | network that's most relevant to your request (3 billion active
         | parameters). This makes the model run faster, especially if you
         | don't have enough VRAM for the whole thing.
         | 
         | The performance/intelligence is said to be about the same as
         | the geometric mean of the total and active parameter counts.
         | So, this model should be equivalent to a dense model with about
         | 10.25 billion parameters.
        
           | zshn25 wrote:
           | Sorry, how did you calculate the 10.25B?
        
             | darrenf wrote:
             | > > _The performance /intelligence is said to be about the
             | same as the geometric mean of the total and active
             | parameter counts. So, this model should be equivalent to a
             | dense model with about 10.25 billion parameters._
             | 
             | > _Sorry, how did you calculate the 10.25B?_
             | 
             | The geometric mean of two numbers is the square root of
             | their product. Square root of 105 (35*3) is ~10.25.
        
           | wongarsu wrote:
           | And even if you have enough VRAM to fit the entire thing,
           | inference speed after the first token is proportional to
           | (activated parameters)/(vram bandwidth)
           | 
           | If you have the vram to spare, a model with more total params
           | but fewer activated ones can be a very worthwhile tradeoff.
           | Of course that's a big if
        
       | aliljet wrote:
       | I'm broadly curious how people are using these local models.
       | Literally, how are they attaching harnesses to this and finding
       | more value than just renting tokens from Anthropic of OpenAI?
        
         | marssaxman wrote:
         | I used vLLM and qwen3-coder-next to batch-process a couple
         | million documents recently. No token quota, no rate limits,
         | just 100% GPU utilization until the job was done.
        
         | Panda4 wrote:
         | I was thinking the same thing. My only guess is that they are
         | excited about local models because they can run it cheaper
         | through Open Router ?
        
         | flux3125 wrote:
         | They are okay for vibe coding throw-away projects without
         | spending your Anthrophic/OAI tokens
        
         | lkjdsklf wrote:
         | The people i know that use local models just end up with both.
         | 
         | The local models don't really compete with the flagship labs
         | for most tasks
         | 
         | But there are things you may not want to send to them for
         | privacy reasons or tasks where you don't want to use tokens
         | from your plan with whichever lab. Things like openclaw use a
         | ton of tokens and most of the time the local models are totally
         | fine for it (assuming you find it useful which is a whole
         | different discussion)
        
           | deaux wrote:
           | The open weights models absolutely compete with flagship labs
           | for most tasks. OpenAI and Anthropic's "cheap tier" models
           | are completely uncompetitive with them for "quality / $" and
           | it's not close. Google is the only one who has remained
           | competitive in the <$5/1M output tier with Flash, and now has
           | an incredibly strong release with Gemma 4.
           | 
           | Unless you have a corporate lock-in/compliance need, there
           | has been no reason to use Haiku or GPT mini/nano/etc over
           | open weights models for a long time now.
        
         | bildung wrote:
         | The privacy/data security angle really is important in some
         | regions and industries. Think European privacy laws or
         | customers demanding NDAs. The value of Anthropic and OpenAI is
         | zero for both cases, so easy to beat, despite local models
         | being dumber and slower.
        
         | oompydoompy74 wrote:
         | Idk about everyone else, but I don't want to rent tokens
         | forever. I want a self hosted model that is completely private
         | and can't be monitored or adulterated without me knowing. I use
         | both currently, but I am excited at the prospect of maybe not
         | having to in the near to mid future.
         | 
         | I've increasingly started self hosting everything in my home
         | lately because I got tired of SAAS rug pulls and I don't see
         | why LLM's should eventually be any different.
        
         | seemaze wrote:
         | Qwen3.5-9B has been extremely useful for local fuzzy table
         | extraction OCR for data that cannot be sent to the cloud.
         | 
         | The documents have subtly different formatting and layout due
         | to source variance. Previously we used a large set of
         | hierarchical heuristics to catch as many edge cases as we could
         | anticipate.
         | 
         | Now with the multi-modal capabilities of these models we can
         | leverage the language capabilities along side vision to extract
         | structured data from a table that has 'roughly this shape' and
         | 'this location'.
        
         | kamranjon wrote:
         | I use LMStudio to host and run GLM 4.7 Flash as a coding agent.
         | I use it with the Pi coding agent, but also use it with the Zed
         | editor agent integrations. I've used the Qwen models in the
         | past, but have consistently come back to GLM 4.7 because of its
         | capabilities. I often use Qwen or Gemma models for their vision
         | capabilities. For example, I often will finish ML training
         | runs, take a photo of the graphs and visualizations of the run
         | metrics and ask the model to tell me things I might look at
         | tweaking to improve subsequent training runs. Qwen 3.5 0.8b is
         | pretty awesome for really small and quick vision tasks like
         | "Give me a JSON representation of the cards on this page".
        
         | Aurornis wrote:
         | It's easy to find a combination of llama.cpp and a coding tool
         | like OpenCode for these. Asking an LLM for help setting it up
         | can work well if you don't want to find a guide yourself.
         | 
         | > and finding more value than just renting tokens from
         | Anthropic of OpenAI?
         | 
         | Buying hardware to run these models is not cost effective. I do
         | it for fun for small tasks but I have no illusions that I'm
         | getting anything superior to hosted models. They can be useful
         | for small tasks like codebase exploration or writing simple
         | single use tools when you don't want to consume more of your
         | 5-hour token budget though.
        
           | toxik wrote:
           | Oh lord, are the LLMs already replacing LLMs?
        
         | znnajdla wrote:
         | Some tasks don't require SOTA models. For translating small
         | texts I use Gemma 4 on my iPhone because it's faster and better
         | than Apple Translate or Google Translate and works offline.
         | Also if you can break down certain tasks like JSON healing into
         | small focused coding tasks then local models are useful
        
           | kaliqt wrote:
           | Is it really better? In which languages?
        
             | deaux wrote:
             | Yes it is and has been for a very long time, it has been
             | years now. Gemini 1.5 Pro is when LLM translations started
             | significantly outperforming non-LLM machine translation,
             | and that came out over 2 years ago.
             | 
             | Ever since then Google models have been the strongest at
             | translation across the board, so it's no surprise Gemma 4
             | does well. Gemini 3 Flash is better at translation than any
             | Claude or GPT model. OpenAI models have always been weakest
             | at it, continuing to this day. It's quite interesting how
             | these characteristics have stayed stable over time and many
             | model versions.
             | 
             | I'm primarily talking about non-trivial language pairs,
             | something like English<>Spanish is so "easy" now it's hard
             | to distinguish the strong models.
        
             | homebrewer wrote:
             | I've been using gemma4 for translating Mongolian to
             | English. It runs circles around Google Translate for that
             | language pair, it's not even close.
        
         | deaux wrote:
         | While they can be run locally, and most of the discussion on HN
         | about that, I bet that if you look at total tok/day local usage
         | is a tiny amount compared to total cloud inference even for
         | these models. Most people who do use them locally just do a
         | prompt every now and then.
        
           | zozbot234 wrote:
           | This is why I'd like to see a lot more focus on batched
           | inference with lower-end hardware. If you just do a tiny
           | amount of tok/day and can wait for the answer to be computed
           | overnight or so, you don't really need top-of-the-line
           | hardware even for SOTA results.
        
             | deaux wrote:
             | > If you just do a tiny amount of tok/day and can wait for
             | the answer to be computed overnight or so
             | 
             | But they can't? The usage pattern is the polar opposite.
             | Most people running these models locally just ask a few
             | questions to it throughout the day. They want the answers
             | now, or at least within a minute.
        
               | zozbot234 wrote:
               | If you want the answer right now, that alone ups your
               | compute needs to the point where you're probably better
               | off just using a free hosted-AI service. Unless the
               | prompt is trivial enough that it can be answered quickly
               | by a tiny local model.
        
         | dist-epoch wrote:
         | There are really nice GUIs for LLMs - CherryStudio for example,
         | can be used with local or cloud models.
         | 
         | There are also web-UIs - just like the labs ones.
         | 
         | And you can connect coding agents like Codex, Copilot or Pi to
         | local coding agents - the support OpenAI compatible APIs.
         | 
         | It's literally a terminal command to start serving the model
         | locally and you can connect various things to it, like Codex.
        
         | kylehotchkiss wrote:
         | I am working on a research project to link churches from their
         | IRS Exempt org BMF entry to their google search result from 10
         | fetched. Gwen2.5-14b on a 16gb Mac Mini. It works good enough!
         | 
         | It's entertaining to see HN increasingly consider coding
         | harness as the only value a model can provide.
        
         | jwitthuhn wrote:
         | I've been largely using Qwen3.5-122b at 6 bit quant locally for
         | some c++/go/python dev lately because it is quite capable as
         | long as I can give it pretty specific asks within the codebase
         | and it will produce code that needs minimal massaging to fit
         | into the project.
         | 
         | I do have a $20 claude sub I can fall back to for anything qwen
         | struggles with, but with 3.5 I have been very pleased with the
         | results.
        
           | 3836293648 wrote:
           | How much VRAM do you need for that?
        
       | tristor wrote:
       | I'm disappointed they didn't release a 27B dense model. I've been
       | working with Qwen3.5-27B and Qwen3.5-35B-A3B locally, both in
       | their native weights and the versions the community distilled
       | from Opus 4.6 (Qwopus), and I have found I generally get higher
       | quality outputs from the 27B dense model than the 35B-A3B MOE
       | model. My basic conclusion was that MoE approach may be more
       | memory efficient, but it requires a fairly large set of active
       | parameters to match similarly sized dense models, as I was able
       | to see better or comparable results from Qwen3.5-122B-A10B as I
       | got from Qwen3.5-27B, however at a slower generation speed. I am
       | certain that for frontier providers with massive compute that MoE
       | represents a meaningful efficiency gain with similar quality, but
       | for running models locally I still prefer medium sized dense
       | models.
       | 
       | I'll give this a try, but I would be surprised if it outperforms
       | Qwen3.5-27B.
        
         | adrian_b wrote:
         | You are right, but this is just the first open-weights model of
         | this family.
         | 
         | They said that they will release several open-weights models,
         | though there was an implication that they might not release the
         | biggest models.
        
           | tristor wrote:
           | I'm totally fine with that, frankly. I'm blessed with 128GB
           | of Unified Memory to run local models, but that's still tiny
           | in comparison the larger frontier models. I'd much rather get
           | a full array of small and medium sized models, and building
           | useful things within the limits of smaller models is more
           | interesting to me anyway.
        
           | hnfong wrote:
           | Given that DeepSeek, GLM, Kimi etc have all released large
           | open weight models, I am personally grateful that Qwen fills
           | the mid/small sized model gap even if they keep their largest
           | models to themselves. The only other major player in the
           | mid/small sized space at this point is pretty much only
           | Gemma.
        
         | ilaksh wrote:
         | It's a given that the dense models with comparable size are
         | better. I also proved that in my use case for those two Qwen
         | 3.5 models.
         | 
         | The benchmarks show 3.6 is a bit better than 3.5. I should
         | retry my task, but I don't have a lot of confidence. But it
         | does sound like they worked on the right thing which is getting
         | closer to the 27B performance.
        
       | seemaze wrote:
       | Fingers crossed for mid and larger models as well. I'd personally
       | love to see Qwen3.6-122B-A10B.
        
         | Vespasian wrote:
         | That would be really great. Though 3.5 122B is already doing a
         | lot of work in our setup.
        
       | nurettin wrote:
       | I tried the car wash puzzle:
       | 
       | You want to wash your car. Car wash is 50m away. Should you walk
       | or go by car?
       | 
       | > Walk. At 50 meters, the round trip is roughly 100 meters,
       | taking about two minutes on foot. Driving would require starting
       | the engine, navigating, parking, and dealing with unnecessary
       | wear for a negligible distance. Walk to the car wash, and if the
       | bay requires the vehicle inside, have it moved there or return on
       | foot. Walking is faster and more efficient.
       | 
       | Classic response. It was really hard to one shot this with
       | Qwen3.5 Q4_K_M.
       | 
       | Qwen3.6 UD-IQ4_XS also failed the first time, then I added this
       | to the system prompt:
       | 
       | > Double check your logic for errors
       | 
       | Then I created a new dialog and asked the puzzle and it
       | responded:
       | 
       | > Drive it. The car needs to be present to be washed. 50 meters
       | is roughly a 1-minute walk or a 10-second drive. Walking leaves
       | the car behind, making the wash impossible. Driving it the short
       | distance is the only option that achieves the goal.
       | 
       | Now 3.6 gets it right every time. So not as great as a super
       | model, but definitely an improvement.
        
         | dist-epoch wrote:
         | Interestingly, Gemma4-26B IQ4_XS gets it correct:
         | 
         | > This sounds like a logic riddle! The answer is: You should go
         | by car. Here is why: If you walk, you will arrive at the car
         | wash, but your car will still be 50 meters away at home. You
         | can't wash the car if the car isn't there! To accomplish your
         | goal, you have to drive the car to the car wash.
         | 
         | It has the wrong one in thinking. It did think longer than
         | usual:
         | 
         | Direct answer: Walk.
         | 
         | Reasoning 1: Distance (50m is negligible).
         | 
         | Reasoning 2: Practicality/Efficiency (engine wear/fuel).
         | 
         | Reasoning 3: Time (walking is likely faster or equal when
         | considering car prep).
         | 
         | ...
         | 
         | Wait, if I'm washing the car, I need to get the car to the car
         | wash. The question asks how I should get there.
         | 
         | ...
         | 
         | Wait, let's think if there's a trick. If you "go by car," you
         | are moving the car to the destination. If you "walk," you are
         | just moving yourself.
         | 
         | Conclusion: You should drive the car.
        
       | lopsotronic wrote:
       | Dangit, I'll need to give this a run on my personal machine. This
       | looks impressive.
       | 
       | At the time of writing, all deepseek or qwen models are de facto
       | prohibited in govcon, including local machine deployments via
       | Ollama or similar. Although no legislative or executive mandate
       | yet exists [1], it's perceived as a gap [2], and contracts are
       | already including language for prohibition not just in the
       | product but any part of the software environment.
       | 
       | The attack surface for a (non-agentic) model running in local
       | ollama is basically non-existent . . but, eh . . I do get it, at
       | some level. While they're not l33t haXX0ring your base, the
       | models are still largely black boxes, can move your attention
       | away from things, or towards things, with no one being the wiser.
       | "Landing Craft? I see no landing craft". This would boil out in
       | test, ideally, but hey, now you know how much time your typical
       | defense subcon spends in meaningful software testing[3].
       | 
       | [1] See also OMB Memorandum M-25-22 (preference for AI developed
       | and produced in the United States), NIST CAISI assessment of PRC-
       | origin AI models as "adversary AI" (September 2025), and House
       | Select Committee on the CCP Report (April 16, 2025), "DeepSeek
       | Unmasked".
       | 
       | [2] Overall, rather than blacklist, I'd recommend a "whitelist"
       | of permitted models, maintained dynamically. This would operate
       | the same way you would manage libraries via SSCG/SSCM (software
       | supply chain governance/management) . . but few if any defense
       | subcons have enough onboard savvy to manage SSCG let alone
       | spooling a parallel construct for models :(. Soooo . . ollama
       | regex scrubbing it is.
       | 
       | [3] i.e. none at all, we barely have the ability to MAKE anything
       | like software, given the combination of underwhelming pay scales
       | and the fact defense companies always seem to have a requirement
       | for on-site 100% in some random crappy town in the middle of BFE.
       | If it wasn't for the downturn in tech we wouldn't have anyone
       | useful at all, but we snagged some silcon refugees.
        
       | alecco wrote:
       | Related interesting find on Qwen.
       | 
       | "Qwen's base models live in a very exam-heavy basin - distinct
       | from other base models like llama/gemma. Shown below are the
       | embeddings from randomly sampled rollouts from ambiguous initial
       | words like "The" and "A":"
       | 
       | https://xcancel.com/N8Programs/status/2044408755790508113
        
       | Glemllksdf wrote:
       | I tried Gemma 4 A4B and was surprised how hart it is to use it
       | for agentic stuff on a RTX 4090 with 24gb of ram.
       | 
       | Balancing KV Cache and Context eating VRam super fast.
        
       | yieldcrv wrote:
       | Anybody use these instead of codex or claude code? Thoughts in
       | comparison?
       | 
       | benchmarks dont really help me so much
        
         | 3836293648 wrote:
         | In my test case (a feature all models got stuck on a few months
         | ago) it just gets stuck in a thinking loop and never gets
         | anywhere. Not a super amazing test, but it happened a few times
         | in a row, so...
        
       | 999900000999 wrote:
       | Looking to move off ollama on Open Suse tumbleweed.
       | 
       | Should I use brew to install llma.ccp or the zypper to install
       | the tumbleweed package?
        
         | rexreed wrote:
         | Why are you looking to move off Ollama? Just curious because
         | I'm using Ollama and the cloud models (Kimi 2.5 and Minimax
         | 2.7) which I'm having lots of good success with.
        
           | 999900000999 wrote:
           | Ollama co mingles online and local models which defeats the
           | purpose for me
        
             | rexreed wrote:
             | You can disable all cloud models in your Ollama settings if
             | you just want all local. For cloud you don't have to use
             | the cloud models unless you explicitly request.
        
         | badsectoracula wrote:
         | You can compile it from source, all you need to do is clone the
         | repository and do a `cmake -B build -DGGML_VULKAN=1` (add other
         | backends if you want) followed by a `cmake --build build
         | --config Release` and then you get all the llama tools in the
         | `build/bin` (including `llama-server` which provides a web-
         | based interface). There is a `docs/build.md` that has more
         | detailed info (especially if you need another backend, though
         | at least on my RX 7900 XTX i see no difference in terms of
         | performance between Vulkan and ROCm and the former is much more
         | stable and compatible -- i tried ROCm for a bit thinking it'd
         | be much faster but only ended up being much more annoying as
         | some models would OOM on it while they worked on Vulkan -- if
         | you or NVIDIA hardware all this may sound quaint though :-P).
        
           | 999900000999 wrote:
           | Cool, I assume this is how adults use llms.
           | 
           | I'm on a nvidia gpu , but I want to be able to combine vram
           | with system memory.
        
         | stratos123 wrote:
         | Why not just download the binaries from github releases?
        
       | amelius wrote:
       | Looks like they compare only to open models, unfortunately.
       | 
       | As I am using mostly the non-open models, I have no idea what
       | these numbers mean.
        
       | psim1 wrote:
       | (Please don't downvote - serious question) Are Chinese models
       | generally accepted for use within US companies? The company I
       | work for won't allow Qwen.
        
         | kelsey98765431 wrote:
         | In private sector yes. Anything that touches public sector
         | (government) and it starts to be supply chain concerns and they
         | want all american made models
        
         | DiabloD3 wrote:
         | There is a difference between Chinese model and Chinese
         | service.
         | 
         | Your company most likely is banning the use of foreign
         | services, but it wouldn't make sense to ban the model, since
         | the model would be ran locally.
         | 
         | I wouldn't allow my employees to use a foreign service either
         | if my company had specific geographic laws it had to follow
         | (ie, fin or med or privacy laws, such as the ones in the EU).
         | 
         | That said, I'm not sure I'd allow them to use any AI product
         | either, locally inferred on-prem or not: I need my employees to
         | _not_ make mistakes, not automate mistake making.
        
       | syntaxing wrote:
       | Is it worth running speculative decoding on small active models
       | like this? Or does MTP make speculative decoding unnecessary?
        
       | andy_ppp wrote:
       | Do we know if other models have started detecting and poisoning
       | training/fine tuning that these Chinese models seem to use for
       | alignment, I'd certainly be doing some naughty stuff to keep my
       | moat if I was Anthropic or OpenAI...
        
         | storus wrote:
         | They no longer show reasoning traces and are throttling more
         | aggressively.
        
           | zozbot234 wrote:
           | They never showed full reasoning traces, just post-hoc
           | summaries.
        
             | storus wrote:
             | DeepSeek still shows them, it sometimes says "I am
             | ChatGPT", and Claude sometimes says "I am DeepSeek" so the
             | distillation went both ways.
        
       | KronisLV wrote:
       | I wonder how this one compares to Qwen3 Coder Next (the 80B A3B
       | model), since you'd think that even though it's older, it having
       | more parameters would make it more useful for agentic and
       | development use cases:
       | https://huggingface.co/collections/Qwen/qwen3-coder-next
        
       | solomatov wrote:
       | Did anyone try it and Gemma 4? Does it feel that it's better than
       | Gemma 4?
        
       | simonw wrote:
       | I've been running this on my laptop with the Unsloth 20.9GB GGUF
       | in LM Studio:
       | https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF/blob/mai...
       | 
       | It drew a better pelican riding a bicycle than Opus 4.7 did!
       | https://simonwillison.net/2026/Apr/16/qwen-beats-opus/
        
         | jamwise wrote:
         | I've had some really gnarly SVGs from Claude. Here's what I got
         | after many iterations trying to draw a hand:
         | https://imgur.com/a/X4Jqius
        
           | giantg2 wrote:
           | Probably because all the training material of humans drawing
           | hands are garbage haha.
        
         | slekker wrote:
         | How does it do with the "car wash" benchmark? :D
        
         | danielhanchen wrote:
         | Oh that is pretty good! And the SVG one!
        
         | bertili wrote:
         | It's fascinating that a $999 Mac Mini (M4 32GB) with almost
         | similar wattage as a human brain gets us this far.
        
         | cyclopeanutopia wrote:
         | But that you also gave a win to Qwen on flamingo is pretty
         | outrageous! :)
         | 
         | Tthe right one looks much better, plus adding sunglasses
         | without prompting is not that great. Hopefully it won't add
         | some backdoor to the generated code without asking. ;)
        
           | simonw wrote:
           | I love how the Chinese models often have an unprompted
           | predilection to add flair.
           | 
           | GLM-5.1 added a sparkling earring to a north Virginia opossum
           | the other day and I was delighted:
           | https://simonwillison.net/2026/Apr/7/glm-51/
        
             | monksy wrote:
             | You're running 5.1 locally or hosted?
        
               | simonw wrote:
               | I used that one via OpenRouter.
        
         | jubilanti wrote:
         | I wonder when pelican riding a bicycle will be useless as an
         | evaluation task. The point was that it was something weird
         | nobody had ever really thought about before, not in the
         | benchmarks or even something a team would run internally. But
         | now I'd bet internally this is one of the new Shirley Cards.
        
           | abustamam wrote:
           | Simon has an article on this
           | 
           | https://simonwillison.net/2025/Nov/13/training-for-
           | pelicans-...
        
           | rafaelmn wrote:
           | I mean look at the result where he asked about a unicycle -
           | the model couldn't even keep the spokes inside the wheels -
           | would be rudimentary if it "learned" what it means to draw a
           | bicycle wheel and could transfer that to unicycle.
        
             | duzer65657 wrote:
             | it's the frame that's surprisingly - and consistentnly -
             | wrong. You'd think two triangles would be pretty easy to
             | repro; once you get that the rest is easy. It's not like
             | he's asking "draw a pelican on a four-bar linkage
             | suspension mountainbike..."
        
               | Reddit_MLP2 wrote:
               | This is older, but even humans don't have a great concept
               | of how a bicycle works...
               | https://twistedsifter.com/2016/04/artist-asks-people-to-
               | draw...
        
               | yndoendo wrote:
               | Wouldn't this be more about being capable of mentally
               | remembering how a bicycle looks versus how it works?
               | 
               | This reminds me of Pictionary. [0] Some people are good
               | and some are really bad.
               | 
               | I am really bad a remembering how items look in my head
               | and fail at drawing in Pictionary. My drawing skills are
               | tied to being able to copy what I see.
               | 
               | [0] https://en.wikipedia.org/wiki/Pictionary
        
               | quinnjh wrote:
               | is it possible to have greater success with the
               | specificity? I don't think i ever drew a bike frame
               | properly as a kid despite riding them and understanding
               | the concept of spokes and wheels...
        
           | MagicMoonlight wrote:
           | They'll hardcode it in 4.8, just like they do when they need
           | to "fix" other issues
        
           | amelius wrote:
           | Yeah try it with something else, or e.g. add a tiger to the
           | back seat.
        
           | SwellJoe wrote:
           | Pelicanmaxxing
        
         | prirun wrote:
         | The flamingo on Qwen's unicycle is sitting on the tire, not the
         | seat. That wins because of sunglasses?
        
           | evilduck wrote:
           | Can a benchmark meant as a joke not use a fun interpretation
           | of results? The Qwen result has far better style points. Fun
           | sunglasses, a shadow, a better ground, a better sky, clouds,
           | flowers, etc.
           | 
           | If we want to get nitty gritty about the details of a joke, a
           | flamingo probably couldn't physically sit on a unicycle's
           | seat and also reach the pedals anyways.
        
           | akavel wrote:
           | Well, maybe the flamingo is a _really good_ unicyclist...
           | 
           | https://youtu.be/Rrpgd5oIKwI
        
         | culi wrote:
         | the more I look at these images the more convinced I become
         | that world models are the major missing piece and that these
         | really are ultimately just stochastic sentence machines. Maybe
         | Chomsky was right
        
         | MeteorMarc wrote:
         | Interesting, qwen has the pelican driving on the left lane.
         | Coincidence or has it something to do with the workers
         | providing the RL data?
        
           | rubiquity wrote:
           | Could be on a bike path where bikes are on the left and
           | pedestrians to the right.
        
         | rdslw wrote:
         | interesting, I just tried this very model, unsloth, Q8, so in
         | theory more capable than Simon's Q4, and get those three
         | "pelicans". definitely NOT opus quality. lmstudio, via Simon's
         | llm, but not apple/mlx. Of course the same short prompt.
         | 
         | Simon, any ideas?
         | 
         | https://ibb.co/gFvwzf7M
         | 
         | https://ibb.co/dYHRC3y
         | 
         | https://ibb.co/FLc6kggm (tried here temperature 0.7 instead of
         | pure defaults)
        
         | kelnos wrote:
         | I'm not sure how you can give the flamingo win to Qwen:
         | 
         | * It's sitting on the tire, not the seat.
         | 
         | * Is that weird white and black thing supposed to be a beak? If
         | so, it's sticking out of the side of its face rather than the
         | center.
         | 
         | * The wheel spokes are bizarre.
         | 
         | * One of the flamingo's legs doesn't extend to the pedal.
         | 
         | * If you look closely at the sunglasses, they're semi-
         | transparent, and the flamingo only has one eye! Or the other
         | eye is just on a different part of its face, which means the
         | sunglasses aren't positioned correctly. Or the other eye isn't.
         | 
         | * (subjective) The sunglasses and bowtie are cute, but you
         | didn't ask for them, so I'd actually _dock_ points for that.
         | 
         | * (subjective) I guess flamingos have multiple tail feathers,
         | but it looks kinda odd as drawn.
         | 
         | In contrast, Opus's flamingo isn't as detailed or fancy, but
         | more or less all of it looks correct.
        
           | withinboredom wrote:
           | He literally said it came down to the comment in the SVG.
           | Points for taste, not correctness. Basically.
        
         | monksy wrote:
         | Hey I really enjoy your blog. On some things I end up finding a
         | blog post of yours thats a year+ old and at other times, you
         | and I are investigating similar things. I just pulled Qwen3.6 -
         | 35b -A3B (Can't believe thats a A3B coming from 35b).
         | 
         | I'm impressed about the reach of your blog, and I'm hoping to
         | get into blogging similar things. I currently have a lot on my
         | backlog to blog about.
         | 
         | In short, keep up the good work with an interesting blog!
        
         | bwv848 wrote:
         | I've been trying the Q4_K_M version, and sometimes it gets
         | stuck in a loop. Gemma 4 doesn't have this issue.
        
           | yencabulator wrote:
           | This has happened before with quantizations and other
           | backends (ones not used by the research lab). Give it a week,
           | download latest versions of everything, and try again.
        
         | GistNoesis wrote:
         | Thanks for pointing to the GGUF.
         | 
         | I just tried this GGUF with llama.cpp in its UD Q4_K_XL version
         | on my custom agentic oritened task consisiting of wiki
         | exploration and automatic database building (
         | https://github.com/GistNoesis/Shoggoth.db/ )
         | 
         | I noted a nice improvement over QWen3.5 in its ability to
         | discover new creatures in the open ended searching task, but
         | I've not quantified it yet with numbers. It also seems faster,
         | at around 140 token/s compared to 100 token/s , but that's
         | maybe due to some different configuration options.
         | 
         | Some little difference with QWen3.5 : to avoid crashes due to
         | lack of memory in multimodal I had to pass --no-mmproj-offload
         | to disable the gpu offload to convert the images to tokens
         | otherwise it would crash for high resolutions images. I also
         | used quantized kv store by passing -ctk q8_0 -ctv q8_0 and with
         | a ctx-size 150000 it only need 23099 MiB of device memory which
         | means no partial RAM offloading when I use a RTX 4090.
        
         | jaspanglia wrote:
         | The real question is what the next truly weird, un-optimized
         | prompt will be. Something involving a sloth debugging a quantum
         | computer in MS Paint?"
        
         | realityfactchex wrote:
         | Here's a reproduction (LM Studio, same Qwen3.6-35B-A3B-GGUF
         | model as linked in parent).
         | 
         | https://files.catbox.moe/uqu0ae.png
         | 
         | - My Qwen 3.6 result had sun and cloud sky similar to the
         | second Opus 4.7 result in Simon's post.
         | 
         | - My Qwen 3.6 result had no grass, but all three results in
         | Simon's post had grass.
         | 
         | - My Qwen 3.6 result had visible "tailing air motion" like
         | Simon's Qwen 3.6 result.
         | 
         | My Qwen 3.6 result had nice "sun with halo" effect that none of
         | Simon's results had.
         | 
         | But, I know, it's more about the pelican and the bicycle.
        
       | tmaly wrote:
       | What is the min VRAM this can run on given it is MOE?
        
         | mncharity wrote:
         | Fwiw, with its predecessor's Qwen3.5-35B-A3B-Q6_K.gguf, on a
         | laptop's 6 GB VRAM and 32 GB RAM, with default llama.cpp
         | settings, I get 20 t/s generation.
        
           | rubiquity wrote:
           | Have you tried running llama.cpp with Unified Memory
           | Access[1] so your iGPU can seamlessly grab some of the RAM?
           | The environment variable is prefixed with CUDA but this is
           | not CUDA specific. It made a pretty significant difference (>
           | 40% tg/s) on my Ryzen 7840U laptop.
           | 
           | 1 - https://github.com/ggml-
           | org/llama.cpp/blob/master/docs/build...
        
             | zozbot234 wrote:
             | Your link seems to be describing a runtime environment
             | variable, it doesn't need a separate build from source. I'm
             | not sure though (1) why this info is in build.md which
             | should be specific to the building process, rather than
             | some separate documentation; and (2) if this really isn't
             | CUDA-specific, why the canonical GGML variable name isn't
             | GGML_ENABLE_UNIFIED_MEMORY , with the _CUDA_ variant
             | treated as a legacy alias. AIUI, both of these should be
             | addressed with pull requests for llama.cpp and/or the ggml
             | library itself.
        
               | rubiquity wrote:
               | You are right that it is an environment variable, and
               | that's how I have it set in my nix config. Thanks for
               | correcting that.
               | 
               | Unfortunately llama.cpp is somewhat notorious for having
               | lackluster docs. Most of the CLI tools don't even tell
               | you what they are for.
        
               | mncharity wrote:
               | Hmm. Perhaps there's a niche for a "The Missing Guide to
               | llama.cpp"? Getting started, I did things like wrapping
               | llama-cli in a pty... and only later noticing a --simple-
               | io argument. I wonder if "living documents" are a thing
               | yet, where LLMs keep an eye on repo and fora, and update
               | a doc autonomously.
        
             | mncharity wrote:
             | I hadn't tried that, thanks! I found simply defining
             | GGML_CUDA_ENABLE_UNIFIED_MEMORY, whether 1, 0, or "", was a
             | 10x hit to 2 t/s. Perhaps because the laptop's RAM is
             | already so over-committed there. But with the much smaller
             | 4B Qwen3.5-4B-Q8_0.gguf, it doubled performance from 20 to
             | 40+ t/s! Tnx! (an old Quadro RTX 3000 rather than an iGPU)
        
       | cyrialize wrote:
       | My last laptop was a used 2012 T530.
       | 
       | My current is a used M1 MBP Pro with 16GB of ram.
       | 
       | I thought this was all I was ever going to need, but wanting to
       | run really nice models locally has me thinking about upgrading.
       | 
       | Although, part of me wants to see how far I could get with my
       | trusty laptop.
        
         | bigyabai wrote:
         | Your current laptop is still a fine thin client. Unless you
         | program in the woods, it's probably cheapest to build a home
         | inference box and route it over Tailscale or something.
        
           | system2 wrote:
           | Or just an API server for all other devices to connect and do
           | stuff with it.
        
       | giantg2 wrote:
       | I cant wait to see some smaller sizes. I would love to run some
       | sort of coding centric agent on a local TPU or GPU instead of
       | having to pay, even if it's slower.
        
       | ActorNightly wrote:
       | Can anyone confirm this fits on a 3090? Size is exactly 24gb
        
       | cpburns2009 wrote:
       | Anyone else getting gibberish when running
       | unsloth/Qwen3.6-35B-A3B-GGUF:UD-IQ4_XS on CUDA (llama.cpp b8815)?
       | UD-Q4_K_XL is fine, as is Vulkan in general.
        
         | cpburns2009 wrote:
         | Apparently it's a known issue with CUDA 13.2 [1].
         | 
         | [1]
         | https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF/discussi...
        
       | codeugo wrote:
       | Are we going to get to the point where a local model can do
       | almost what sonnet 4.6 can do?
        
         | bluerooibos wrote:
         | Of course we are. And Opus 4.6+. It's a matter of when, not if.
        
         | intothemild wrote:
         | We're already there IMHO.. If you have enough ram, sure.. but
         | the ~32gig people can run models that beat sonnet 4.5
        
       | the__alchemist wrote:
       | Is this the hybrid variant of Gwent and Quen? I hope this is in
       | The Witcher IV!
        
       | zengid wrote:
       | any tips for running it locally within an agent harness? maybe
       | using pi or opencode?
        
         | stratos123 wrote:
         | It pretty much just works. Run the unsloth quant in llama.cpp
         | and hook it up to pi. A bunch of minor annoyances like not
         | having support for thinking effort. It also defaults to
         | "interleaved thinking" (thinking blocks get stripped from
         | context), set `"chat_template_kwargs": {"preserve_thinking":
         | True},` if you interrupt the model often and don't want it to
         | forget what it was thinking.
        
       | kanemcgrath wrote:
       | I have been using Qwen3.5-35B-A3B a lot in local testing, and it
       | is by far the most capable model that could fit on my machine. I
       | think quantization technology has really upped its game around
       | these models, and there were two quants that blew me away
       | 
       | Mudler APEX-I-Quality. then later I tried Byteshape
       | Q3_K_S-3.40bpw
       | 
       | Both made claims that seemed too good to be true, but I couldn't
       | find any traces of lobotomization doing long agent coding loops.
       | with the byteshape quant I am up to 40+ t/s which is a speed that
       | makes agents much more pleasant. On an rtx 3060 12GB and 32GB of
       | system ram, I went from slamming all my available memory to
       | having like 14GB to spare.
        
       | 3836293648 wrote:
       | Qwen3.6 and Gemma4 have the same issue of never getting to the
       | point and just getting stuck in never ending repeating thought
       | loops. Qwen3.5 is still the best local model that works.
        
         | agentifysh wrote:
         | I think the hype around Qwen and even Gemma4 often floated for
         | views/attention glosses over that these models have clear gaps
         | behind what closed models offer.
         | 
         | In short, it has its uses but it would/should not be the main
         | driver. Will it get better, I'm sure of it, but there is too
         | much hype and exaggeration over open source models, for one the
         | hardware simply isn't enough at a price point where we can run
         | something that can seriously compete with today's closed
         | models.
         | 
         | If we got something like GPT-5.4-xhigh that can run on some
         | local hardware under 5k, that would be a major milestone.
        
       | tech_curator wrote:
       | does this run on CPUs as well? Anyone faced any issues? or do you
       | prefer to run using APIs from model providers and aggregators
       | such as openrouter, qubrid etc
        
       ___________________________________________________________________
       (page generated 2026-04-16 23:00 UTC)