[HN Gopher] GLM5 Released on Z.ai Platform
___________________________________________________________________
GLM5 Released on Z.ai Platform
Author : CuriouslyC
Score : 205 points
Date : 2026-02-11 13:42 UTC (9 hours ago)
(HTM) web link (chat.z.ai)
(TXT) w3m dump (chat.z.ai)
| cl0ckt0wer wrote:
| Soft launch? I can't find a blog post on their website.
| cl0ckt0wer wrote:
| The second sentence from a creative writing prompt:
|
| Valerius stood four meters tall--roughly thirteen feet. He was
| not merely a Space Marine; he was a biological singularity.
|
| I'm surprised they still have the emdash and "not x, but y"
| quirks
| htrp wrote:
| distillation is a hell of a drug
| mythz wrote:
| They announced it on twitter [1]:
|
| > A new model is now available on http://chat.z.ai.
|
| Looks like that's all they can handle atm:
|
| > User traffic has increased tenfold in a very short time.
| We're currently scaling to handle the load.
|
| [1] https://x.com/Zai_org/status/2021564343029203032
| zozbot234 wrote:
| There was a one-line X post about something new being available
| at their chat endpoint, but that's about it at the time of this
| writing. Nothing at GitHub or HuggingFace, no tech report or
| anything.
| linolevan wrote:
| What's funny is it's available on /v1/models, but if you call
| it you get an error saying it's not accessible yet. No word
| on pricing, probably the same as 4.7 if I had to guess
| (0.6/2.2)
| Mashimo wrote:
| Wut? Was glm 4.7 not just a few weeks ago?
|
| I wonder if I will be able to use it with my coding plan. Paid
| just 9 usd for 3 month.
| ExpertAdvisor01 wrote:
| Nope. Lite plan receives only same-tier model updates.
| Havoc wrote:
| I don't see it as selectable my side either (opencode & max
| plan)
| ExpertAdvisor01 wrote:
| They updated it now
| Havoc wrote:
| No luck here. Did you do anything specific to make it
| show / reauth or something?
|
| ah nvm - found the guidance on how to change it
| _joel wrote:
| It's avaiable in mine, I think I paid about the same
| NitpickLawyer wrote:
| > It's avaiable in mine
|
| Weird, mine (lite plan) says "Only supports GLM-4.7, GLM-4.6,
| GLM-4.5, and GLM-4.5-Air" and "Get same-tier model updates"
| ...
| Mashimo wrote:
| It seems like nothing is updated yet, expect the chat.
| https://z.ai/subscribe
|
| It all just mentions 4.7
|
| Seems like time will tell.
|
| Edit: They updated it:
|
| > The Lite / Pro plan currently does not include GLM-5
| quota (we will gradually expand the scope and strive to
| enable more users to experience and use GLM-5). If you call
| GLM-5 under the plan endpoints, an error will be returned.
| If you still wish to experience GLM-5 at this stage and are
| willing to pay according to the Pricing, you can call it
| through the General API endpoint (i.e.,
| https://api.z.ai/api/paas/v4/chat/completions), with the
| deduction priority being [Platform Credits - Account
| Balance] in sequence.
| AntonyGarand wrote:
| The documentation is not updated, but it works if you
| hardcode the model id to `GLM-5` within your tool
| NitpickLawyer wrote:
| Cool, thanks. Did you try it out, how's the performance?
| I saw on openrouter that the stealth model was served at
| ~19t/s. Is it any better on their endpoints?
| cmrdporcupine wrote:
| GLM 4.7 _Flash_ was just a few weeks ago. 4.7 fully I think was
| a ways further back early December?
| alias_neo wrote:
| What's the use case for Zai/GLM? I'm currently on Claude Pro,
| and the Zai looks about 50% more expensive after the first 3
| months and according to their chart GLM 4.7 is not quite as
| capable as Opus 4.5?
|
| I'm looking to save on costs because I use it so infrequently,
| but PAYG seems like it'd cost me more in a single session per
| month than the monthly cost plan.
| BeetleB wrote:
| If you pay for the whole year, GLM4.7 is only $7/mo for the
| first year. And until a few days ago, they had a fantastic
| deal that ran for almost 2 months where it was less than
| $3/mo for the first year. I grabbed it, and have been using
| it exclusively for personal coding since. It's good enough
| for me.
|
| The other claimed benefit is a higher quota of tokens.
| Mashimo wrote:
| > What's the use case for Zai/GLM?
|
| It's cheap :) It seems they stopped it now, but for the last
| 2 month you could buy the lite plan for a whole year for
| under 30 USD, while claude is ~19 USD per month. I bought 3
| month for ~9 USD.
|
| I use it for hobby projects. Casual coding with Open Code.
|
| If price is not important Opus / Codex are just plain better.
| mythz wrote:
| It's looking like we'll have Chinese OSS to thank for being able
| to host our own intelligence, free from the whims of proprietary
| megacorps.
|
| I know it doesn't make financial sense to self-host given how
| cheap OSS inference APIs are now, but it's comforting not being
| beholden to anyone or requiring a persistent internet connection
| for on-premise intelligence.
|
| Didn't expect to go back to macOS but they're basically the only
| feasible consumer option for running large models locally.
| mikrl wrote:
| >I know it doesn't make financial sense to self-host given how
| cheap OSS inference APIs are now
|
| You can calculate the exact cost of home inference, given you
| know your hardware and can measure electrical consumption and
| compare it to your bill.
|
| I have no idea what cloud inference in aggregate actually
| costs, whether it's profitable or a VC infused loss leader that
| will spike in price later.
|
| That's why I'm using cloud inference now to build out my local
| stack.
| mythz wrote:
| Not concerned with electricity cost - I have solar + battery
| with excess supply where most goes back to the grid for $0
| compensation (AU special).
|
| But I did the napkin math on M3 Ultra ROI when DeepSeek V3
| launched: at $0.70/2M tokens and 30 tps, a $10K M3 Ultra
| would take ~30 years of non-stop inference to break even -
| without even factoring in electricity. You clearly don't
| self-host to save money. You do it to own your intelligence,
| keep your privacy, and not be reliant on a persistent
| internet connection.
| NiloCK wrote:
| > Didn't expect to go back to macOS but their basically the
| only feasible consumer option for running large models locally.
|
| I presume here you are referring to running on the device in
| your lap.
|
| How about a headless linux inference box in the closet /
| basement?
|
| Return of the home network!
| Aurornis wrote:
| Apple devices have high memory bandwidth necessary to run
| LLMs at reasonable rates.
|
| It's possible to build a Linux box that does the same but
| you'll be spending a lot more to get there. With Apple, a
| $500 Mac Mini has memory bandwidth that you just can't get
| anywhere else for the price.
| cmrdporcupine wrote:
| But a $500 Mac Mini has nowhere near the memory _capacity_
| to run such a model. You 'd need at least 2 512GB machines
| chained together to run this model. Maybe 1 if you
| quantized the crap out of it.
|
| And Apple completely overcharges for memory, so.
|
| This is a model you use via a cheap API provider like
| DeepInfra, or get on their coding plan. It's nice that it
| will be available as open weights, but not practical for
| mere mortals to run.
|
| But I _can_ see a large corporation that wants to avoid
| sending code offsite setting up their own private infra to
| host it.
| zozbot234 wrote:
| The needed memory capacity depends on active parameters
| (not the same as total with a MoE model) and context
| length for the purpose of KV caching. Even then the KV
| cache can be pushed to system RAM and even farther out to
| swap, since writes to it are small (just one KV vector
| per token).
| zozbot234 wrote:
| And then only Apple devices have 512GB of unified memory,
| which matters when you have to combine larger models (even
| MoE) with the bigger context/KV caching you need for
| agentic workflows. You can make do with less, but only by
| slowing things down a whole lot.
| ingenieroariel wrote:
| With Apple devices you get very fast predictions once it
| gets going but it is inferior to nvidia precisely during
| prefetch (processing prompt/context) before it really gets
| going.
|
| For our code assistant use cases the local inference on
| Macs will tend to favor workflows where there is a lot of
| generation and little reading and this is the opposite of
| how many of use use Claude Code.
|
| Source: I started getting Mac Studios with max ram as soon
| as the first llama model was released.
| storus wrote:
| This. It's awful to wait 15 minutes for M3 Ultra to start
| generating tokens when your coding agent has 100k+ tokens
| in its context. This can be partially offset by adding
| DGX Spark to accelerate this phase. M5 Ultra should be
| like DGX Spark for prefill and M3 Ultra for token
| generation but who know when it will pop up and for how
| much? And it still will be at around 3080 GPU levels just
| with 512GB RAM.
| zozbot234 wrote:
| All Apple devices have a NPU which is potentially able to
| save power for compute bound operations like prefill (at
| least if you're ok with FP16 FMA/INT8 MADD arithmetic).
| It's just a matter of hooking up support to the main
| local AI frameworks. This is not a speedup per se but
| gives you more headroom wrt. power and thermals for
| everything else, so should yield higher performance
| overall.
| d3k wrote:
| AFAIK, only CoreML can use Apple's NPU (ANE). Pytorch,
| MLX and the other kids on the block use MPS (the GPU). I
| think the limitations you mentioned relate to that (but I
| might be missing something)
| Aurornis wrote:
| > With Apple devices you get very fast predictions once
| it gets going but it is inferior to nvidia precisely
| during prefetch (processing prompt/context) before it
| really gets going
|
| I have a Mac and an nVidia build and I'm not disagreeing
|
| But nobody is building a useful nVidia LLM box for the
| price of a $500 Mac Mini
|
| You're also not getting as much RAM as a Mac Studio
| unless you're stacking multiple $8,000 nVidia RTX 6000s.
|
| There is always something faster in LLM hardware. Apple
| is popular for the price points of average consumers.
| FuckButtons wrote:
| Vllm-mlx with prefix caching helps with this.
| ac29 wrote:
| > a $500 Mac Mini has memory bandwidth that you just can't
| get anywhere else for the price.
|
| The cheapest new mac mini is $600 on Apple's US store.
|
| And it has a 128-bit memory interface using LPDDR5X/7500,
| nothing exotic. The laptop I bought last year for <$500 has
| roughly the same memory speed and new machines are even
| faster.
| jsheard wrote:
| > The cheapest new mac mini is $600 on Apple's US store.
|
| And you're only getting 16GB at that base spec. It's
| $1000 for 32GB, or $2000 for 64GB plus the requisite SOC
| upgrade.
|
| > And it has a 128-bit memory interface using
| LPDDR5X/7500, nothing exotic.
|
| Yeah, 128-bit is table stakes and AMD is making 256-bit
| SOCs as well now. Apple's higher end Max/Ultra chips are
| the ones which stand out with their 512 and 1024-bit
| interfaces. _Those_ have no direct competition.
| pja wrote:
| Only the M4 Pro Mac Minis have faster RAM than you'll get
| in an off-the-shelf Intel/AMD laptop. The M4 Pros start at
| $1399.
|
| You want the M4 Max (or Ultra) in the Mac Studios to get
| the real stuff.
| jannniii wrote:
| Indeed and I got two words for you:
|
| Strix Halo
| esafak wrote:
| How much memory does yours have, what are you running on
| it, with what cache size, and how fast?
| SillyUsername wrote:
| Also, cheaper... X99 + 8x DDR4 + 2696V4 + 4x Tesla P4s
| running on llama.cpp. Total cost about $500 including case
| and a 650W PSU, excluding RAM. Running TDP about 200W non
| peak 550W peak (everything slammed, but I've never seen it
| and I've an AC monitor on the socket). GLM 4.5 Air (60GB
| Q3-XL) when properly tuned runs at 8.5 to 10 tokens /
| second, with context size of 8K. Throw in a P100 too and
| you'll see 11-12.5 t/s (still tuning this one). Performance
| doesn't drop as much for larger model sizes as the
| internode communication and DDR4 2400 is the limiter, not
| the GPUs. I've been using this with 4 channel 96GB ram,
| recently updated to 128GB.
| Aurornis wrote:
| > Also, cheaper... X99 + 8x DDR4 + 2696V4 + 4x Tesla P4s
| running on llama.cpp. Total cost about $500 including
| case and a 650W PSU, excluding RAM.
|
| Excluding RAM in your pricing is misleading right now.
|
| That's a lot of work and money just to get 10 tokens/sec
| mythz wrote:
| Not feasible for Large models, it takes 2x M3 512GB Ultra's
| to run the full Kimi K2.5 model at a respectable 24 tok/s.
| Hopefully the M5 Ultra will can improve on that.
| swalsh wrote:
| Yeah that sounds great until it's running as an autonomous
| moltbot in a distributed network semi-offline with access to
| your entire digital life, and China sneaks in some hidden
| training so these agents turn into an army of sleeper agents.
| resters wrote:
| sleeper agents to do what? let's see how far you can take the
| absurd threat porn fantasy. I hope it was hyperbole.
| swalsh wrote:
| I replied to the comment who doubted me in a more polite
| manner.
| falcor84 wrote:
| There was research last year [0] finding significant
| security issues with the Chinese-made Unitree robots,
| apparently being pre-configured to make it easy to
| exfiltrate data via wi-fi or BLE. I know it's not the same
| situation, but at this stage, I wouldn't blame anyone for
| "absurd threat porn fantasy" - the threats are real, and
| present-day agentic AI is getting really good at
| autonomously exploiting vulnerabilities, whether it's an
| external attacker using it, or whether "the call is coming
| from inside the house".
|
| [0] https://spectrum.ieee.org/unitree-robot-exploit
| jfaat wrote:
| Lol wat? I mean you certainly have enough control self
| hosting the model to not let it join some moltbot network...
| or what exactly are you saying would happen?
| swalsh wrote:
| We just saw last week people are setting up moltbots with
| virtually no knowledge of what it has and doesn't have
| access. The scenario that i'm afraid of is China realizes
| the potential of this. They can add training to the models
| commonly used for assistants. They act normal, are helpful,
| everything you'd want a bot to do. But maybe once in a
| while it checks moltbook or some other endpoint China
| controls for a trigger word. When it sees that, it kicks
| into a completely different mode, maybe it writes a script
| to DDoS targets of interest, maybe it mines your email for
| useful information, maybe the user has credentials to some
| piece that is a critical component of an important supply
| chain. This is not a wild scenario, no new sci-fi
| technology would need to be invented. Everything to do it
| is available today, people are configuring it, and using it
| like this today. The part that I fear is if it is running
| locally, you can't just shut off API access and kill the
| threat. It's running on it's own server, it's own model.
| You have to cut off each node.
|
| Big fan of AI, I use local models A LOT. I do think we have
| to take threats like this seriously. I don't Think it's a
| wild scifi idea. Since WW2, civilians have been as much of
| an equal opportunity target as a soldier, war is about
| logistics, and civilians supply the military.
| resters wrote:
| Fair point but I would be more worried about the US
| government doing this kind of thing to act against US
| citizens than the Chinese government doing it.
|
| I think we're in a brief period of relative freedom where
| deep engineering topics can be discussed with AI agents
| even though they have potential uses in weapons systems.
| Imagine asking chat gpt how to build a fertilizer bomb,
| but apply the same censorship to anything related to
| computer vision, lasers, drone coordination, etc.
| saubeidl wrote:
| What if the US government does instead?
|
| I don't consider them more trustworthy at this point.
| tw1984 wrote:
| exactly, we all need to use CIA/NSA approved models to stay
| safe.
|
| very smart idea!
| Aurornis wrote:
| > It's looking like we'll have Chinese OSS to thank for being
| able to host our own intelligence, free from the whims of
| proprietary megacorps.
|
| I don't know where you draw the line between proprietary
| megacorp and not, but Z.ai is planning to IPO soon as a multi
| billion dollar company. If you think they don't want to be a
| multi billion dollar megacorp like all of the other LLM
| companies I think that's a little short sighted. These models
| are open weight, but I wouldn't count them as OSS.
|
| Also Chinese companies aren't the only companies releasing open
| weight models. ChatGPT has released open weight models, too.
| joshstrange wrote:
| > Also Chinese companies aren't the only companies releasing
| open weight models. ChatGPT has released open weight models,
| too.
|
| I was with you until here. The scraps OpenAI has released
| don't really compare to the GLM models or DeepSeek models (or
| others) in both cadence and quality (IMHO).
| gz5 wrote:
| hopefully it will spread - many open options, from many
| entities, globally.
|
| it is brilliant business strategy from China so i expect it to
| continue and be copied - good things.
|
| reminds me of Google's investments into K8s.
| btbuildem wrote:
| > doesn't make financial sense to self-host
|
| I guess that's debatable. I regularly run out of quota on my
| claude max subscription. When that happens, I can sort of kind
| of get by with my modest setup (2x RTX3090) and quantized
| Qwen3.
|
| And this does not even account for privacy and availability.
| I'm in Canada, and as the US is slowly consumed by its spiral
| of self-destruction, I fully expect at some point a digital
| iron curtain will go up. I think it's prudent to have
| alternatives, especially with these paradigm-shattering tools.
| jsheard wrote:
| I think AI may be the only place you could get away with
| calling a 2x350W GPU rig "modest".
|
| That's like ten normal computers worth of power for the GPUs
| alone.
| kataklasm wrote:
| Did you even try to read and understand the parent comment?
| They said they regularly run out of quota on the exact
| subscription you're advising they subscribe to.
| h3half wrote:
| Pot, kettle
| dymk wrote:
| That's maybe a few dollars to tens of dollars in
| electricity per month depending on where in the US you live
| bigyabai wrote:
| > That's like ten normal computers worth of power for the
| GPUs alone.
|
| Maybe if your "computer" in question is a smartphone?
| Remember that the M3 Ultra is a 300w+ chip that won't beat
| _one_ of those 3090s in compute or raster efficiency.
| jsheard wrote:
| I wouldn't class the M3 Ultra as a "normal" computer
| either. That's a big-ass workstation. I was thinking
| along the lines of a typical Macbook or Mac Mini or
| Windows laptop, which are fine for 99% of anyone who
| isn't looking to play games or run gigantic AI models
| locally.
| bigyabai wrote:
| Those aren't "normal" computers, either. They're iPad
| chips running in the TDP envelope of a tablet, usually
| with iPad-level performance to match.
| flaviolivolsi wrote:
| How does your quantized Qwen3 compares in code quality to
| Opus?
| Aurornis wrote:
| Not the person you're responding to, but my experience with
| models up through Qwen3-coder-next is that they're not even
| close.
|
| They can do a lot of simple tasks in common frameworks
| well. Doing anything beyond basic work will just burn
| tokens for hours while you review and reject code.
| btbuildem wrote:
| It's just as fast, but not nearly as clever. I can push the
| context size to 120k locally, but quality of the work it
| delivers starts to falter above say 40k. Generally you have
| to feed it more bite-sized pieces, and keep one chat to one
| topic. It's definitely a step down from SOTA.
| mythz wrote:
| Did the napkin math on M3 Ultra ROI when DeepSeek V3
| launched: at $0.70/2M tokens and 30 tps, a $10K M3 Ultra
| would take ~30 years of non-stop inference to break even -
| without even factoring in electricity. Clearly people aren't
| self-hosting to save money.
|
| I've got a lite GLM sub $72/yr which would require 138 years
| to burn through the $10K M3 Ultra sticker price. Even GLM's
| highest cost Max tier (20x lite) at $720/yr would buy you ~14
| years.
| wongarsu wrote:
| And it's worth noting that you can get DeepSeek at those
| prices from DeepSeek (Chinese), DeepInfra (US with
| Bulgarian founder), NovitaAI (US), AtlasCloud (US with
| Chinese founder), ParaSail (US), etc. There is no shortage
| of companies offering inference, with varying levels of
| trustworthiness, certificates and promises around (lack of)
| data retention. You just have to pick one you trust
| oceanplexian wrote:
| Doing inference with a Mac Mini to save money is more or
| less holding it wrong. Of course if you buy some overpriced
| Apple hardware it's going to take years to break even.
|
| Buy a couple real GPUs and do tensor parallelism and
| concurrent batch requests with vllm and it becomes
| extremely cost competitive to run your own hardware.
| mythz wrote:
| > Doing inference with a Mac Mini to save money is more
| or less holding it wrong.
|
| No one's running these large models on a Mac Mini.
|
| > Of course if you buy some overpriced Apple hardware
| it's going to take years to break even.
|
| Great, where can I find cheaper hardware that can run GLM
| 5's 745B or Kimi K2.5 1T models? Currently it requires 2x
| M3 Ultras (1TB VRAM) to run Kimi K2.5 at 24 tok/s [1]
| What are the better value alternatives?
|
| [1] https://x.com/alexocheema/status/2016404573917683754
| Gracana wrote:
| Six months ago I'd have said EPYC Turin. You could do a
| heck of a build with 12Ch DDR5-6400 and a GPU or two for
| the dense model parts. 20k would have been a huge budget
| for a homelab CPU/GPU inference rig at the time. Now 20k
| won't buy you the memory.
| ljosifov wrote:
| Everyone should do the calculation for themselves. I too
| pay for couple of subs. But I'm noticing having an agent
| work for me 24/7 changes the calculation somewhat. Often
| not taken into account: the price of input tokens. To
| produce 1K of code for me, the agent may need to churn
| through 1M of tokens of codebase. IDK if that will be
| cached by the API provider or not, but that makes x5-7
| times price difference. OK discussion today about that and
| more https://x.com/alexocheema/status/2020626466522685499
| DeathArrow wrote:
| I don't think an Apple PC can run full Deepseek or GLM
| models.
|
| Even if you quantize the hell out of the models to fit in
| the memory, they will be very slow.
| wongarsu wrote:
| Self-hosting training (or gaming) makes a lot of sense, and
| once you have the hardware self-hosting inference on it is an
| easy step.
|
| But if you have to factor in hardware costs self-hosting
| doesn't seem attractive. All the models I can self-host I can
| browse on openrouter and instantly get a provider who can get
| great prices. With most of the cost being in the GPUs
| themselves it just makes more sense to have others do it with
| better batching and GPU utilization
| zozbot234 wrote:
| If you can get near 100% utilization for your own GPUs
| (i.e. you're letting requests run overnight and not
| insisting on any kind of realtime response) it starts to
| make sense. OpenRouter doesn't have any kind of batched
| requests API that would let you leverage that possibility.
| sowbug wrote:
| In Silicon Valley we pay PG&E close to 50 cents per kWh.
| An RTX 6000 PC uses about 1 kW at full load, and renting
| such a machine from vast.ai costs 60 cents/hour as of
| this morning. It's very hard for heavy-load local AI to
| make sense here.
| Imustaskforhelp wrote:
| And you are forgetting the fact that things like vast.ai
| subscriptions would STILL be more expensive than
| Openrouter's api pricing and even more so in the case of
| AI subscriptions which actively LOSE money for the
| company.
|
| So I would still point out the GP (Original comment)
| where yes, it might not make financial sense to run these
| AI Models [They make sense when you want privacy etc,
| which are all fair concerns but just not financial sense]
|
| But the fact that these models are open source still
| means that they can be run when maybe in future the
| dynamics might shift and it might make sense running such
| large models locally. Even just giving this possibility
| and also the fact that multiple providers could now
| compete in say openrouter etc. as well. All facts
| included, definitely makes me appreciate GLM & Kimi
| compared to proprietory counterparts.
|
| Edit: I highly recommend this video a lot
| https://www.youtube.com/watch?v=SmYNK0kqaDI [AI
| subscription vs H100]
|
| This video is honestly one of the best in my opinion
| about this topic that I watched.
| HumanOstrich wrote:
| Why did you quote yourself at the end of this comment?
| Imustaskforhelp wrote:
| Oops sorry. Fixed it now but I am trying a HN progressive
| extension and what it does is if I have any text selected
| it can actually quote it and I think this is what
| might've happened or such a bug I am not sure.
|
| It's fixed now :)
| btbuildem wrote:
| Yikes.. I pay ~7C/ per kWh in Quebec. In the winter the
| inference rig doubles as a space heater for the office, I
| don't feel bad about running local energy-wise.
| spmurrayzzz wrote:
| For inference, even with continuous batching, getting
| 100% MFUs is basically impossible to do in practice. Even
| the frontier labs struggle with this in highly efficient
| infiniband clusters. Its slightly better with training
| workloads just due to all the batching and parallel
| compute, but still mostly unattainable with consumer rigs
| (you spend a lot of time waiting for I/O).
|
| I also don't think the 100% util is necessary either, to
| be fair. I get a lot of value out of my two rigs (2x rtx
| pro 6000, and 4x 3090) even though it may not be 24/7
| 100% MFU. I'm always training, generating datasets,
| running agents, etc. I would never consider this a
| positive ROI measured against capex though, that's not
| really the point.
| zozbot234 wrote:
| Isn't this just saying that your GPU use is bottlenecked
| by things such as VRAM bandwidth and RAM-VRAM transfers?
| That's normal and expected.
| spmurrayzzz wrote:
| No I'm saying there are quite a few more bottlenecks than
| that (I/O being a big one). Even in the more efficient
| training frameworks, there's per-op dispatch overhead in
| python itself. All the boxing/unboxing of python objects
| to C++ handles, dispatcher lookup + setup, all the
| autograd bookkeeping, etc.
|
| All of the bottlenecks in sum is why you'd never get to
| 100% MFUs (but I was conceding you probably don't need to
| in order to get value)
| 7thpower wrote:
| Unless you already had those cards, it probably still doesn't
| make sense from a purely financial perspective unless you
| have other things you're discounting for.
|
| Doesn't mean you shouldn't do it though.
| visarga wrote:
| Your $5,000 PC with 2 GPUs could have bought you 2 years of
| Claude Max, a model much more powerful and with longer
| context. In 2 years you could make that investment back in
| pay raise.
| benterix wrote:
| > In 2 years you could make that investment back in pay
| raise.
|
| Could you elaborate? I fail to grasp the implication here.
| tw1984 wrote:
| > In 2 years you could make that investment back in pay
| raise.
|
| you can't be a happy uber driver making more money in the
| next 24 months by having a fancy car fitted with the best
| FSD in town when all cars in your town have the same FSD.
| visarga wrote:
| But they don't have the same human in the loop though.
| tw1984 wrote:
| that software is called autonomous agents, the term
| autonomous has nothing to do with human in the loop, it
| is the complete opposite.
| dymk wrote:
| This claim has so many assumptions mixed in it's utterly
| useless
| Aurornis wrote:
| > I regularly run out of quota on my claude max subscription.
| When that happens, I can sort of kind of get by with my
| modest setup (2x RTX3090) and quantized Qwen3.
|
| When talking about fallback from Claude plans, The correct
| financial comparison would be the same model hosted on
| OpenRouter.
|
| You could buy a lot of tokens for the price of a pair of
| 3090s and a machine to run them.
| bigyabai wrote:
| > You could buy a lot of tokens for the price of a pair of
| 3090s and a machine to run them.
|
| That's a subjective opinion, to which the answer is "no you
| can't" for many people.
| nialv7 wrote:
| > Didn't expect to go back to macOS but their basically the
| only feasible consumer option for running large models locally.
|
| Framework Desktop! Half the memory bandwidth of M4 Max, but
| much cheaper.
| thebruce87m wrote:
| Does that equate to half the speed in terms of output? Any
| recommended benchmarks to look at?
| nialv7 wrote:
| https://kyuz0.github.io/amd-strix-halo-toolboxes/
| throwaw12 wrote:
| our laptops, devices, phones, equipments, home stuff are all
| powered by Chinese companies.
|
| It wouldn't surprise me if at some point in the future my local
| "Alexa" assistant will be fully powered by local Chinese OSS
| models with Chinese GPUs and RAM.
| andersa wrote:
| They haven't published the weights yet, don't celebrate too
| early.
| andersa wrote:
| Now they have!
| vidarh wrote:
| I don't really care about being able to _self_ host these
| models, but getting to a point where the hosting is
| _commoditised_ so I know I can switch providers on a whim
| matters a great deal.
|
| Of course, it's nice if I can run it myself as a last resort
| too.
| fauigerzigerk wrote:
| _>...free from the whims of proprietary megacorps_
|
| In one sense yes, but the training data is not open, nor is the
| data selection criteria (inclusions/exclusions, censorship,
| safety, etc). So we are still subject to the whims of someone
| much more powerful that ourselves.
|
| The good thing is that open weights models can be finetuned to
| correct any biases that we may find.
| TheRealPomax wrote:
| Not going to call $30/mo for a github copilot subscription
| "cheap". More like "extortionary".
| cmrdporcupine wrote:
| Yeah it's funny how the needle has moved on this kind of
| thing.
|
| Two years ago people scoffed at buying a personal license for
| e.g. JetBrains IDEs which netted out to $120 USD or something
| a year; VS Code etc took off because they were "free"
|
| But now they're dumping monthly subs to OpenAI and Anthropic
| that work out to the same as their car insurance payments.
|
| It's not sustainable.
| TheRealPomax wrote:
| There's also zero incentive for individual companies to
| care: if I only want to use opus in VS code (and why would
| I use anything else, it's so much better at the job) I can
| either pay for copilot, which has excellent VS Code
| integration (because it has to), or I can pay Claude
| specifically and then use their extension which has the
| absolute worst experience because not only is the chat
| "whimsical, to make AI fun!", its interface is pat of the
| sidebar, so it's mutually exclusive with your file browser,
| search, etc.
|
| So whether you pay Claude or GitHub, Claude gets paid the
| same. So the consumer ends up footing a bill that has no
| reason to exist, and has no real competition because open
| source models can't run at the scale of an Opus or ChatGPT.
|
| (not unless the EU decides it's time for a "European Open
| AI Initiative" where any EU citizen gets free access to an
| EU wide datacenter backed large scale system that AI
| companies can pay to be part of, instead of getting paid to
| connect to)
| mminer237 wrote:
| I'm not sure being beholden to the whims of the Chinese
| Communist Party is an iota better than the whims of proprietary
| megacorps, especially given this probably will become part of a
| megacorp anyway.
| hnfong wrote:
| It seems you missed the point entirely once you saw the word
| "Chinese". The point isn't that the models are from China.
| It's that the weights are open. You can download the weights
| and finetune them yourself. Nobody is beholden to anything.
| muyuu wrote:
| you have 128GB strix halo machines for US$ ~3k
|
| these run some pretty decent models locally, currently I'd
| recommend GPT-OSS 120GB, Qwen Coder Next 80B (either Q8 or Q6
| quants, depending on speed/quality trade-offs) and the very
| best model you can run right now which is Step 3.5 Flash
| (ubergarm GGUF quant) with 256K context although this does push
| it to the limit - GLMs and nemotrons also worth trying
| depending on your priorities
|
| there's clearly a big quantum leap in the SotA models using
| more than 512GB VRAM, but i expect that in a year or two, the
| current SotA is achievable with consumer level hardware, if
| nothing else hardware should catch up with running Kimi 2.5 for
| cheaper than 2x 512GB mac studio ultras - perhaps medusa halo
| next year supports 512GB and DDR5 comes down again, and that
| would put a local whatever the best open model of that size is
| next year within reach of under-US$5K hardware
|
| the odd thing is that there isn't much in this whole range
| between 128GB and 512GB VRAM requirement to justify the huge
| premium you pay for Macs in that range - but this can change at
| any point as every other day there are announcements
| saubeidl wrote:
| And you can get Strix Halo in a Laptop that looks and feels
| like a Macbook Pro that can run Linux if you buy an HP ZBook
| G1A.
|
| Super happy with that thing, only real downside is battery
| life.
| pzo wrote:
| AFAIK they haven't released this one as OSS yet. They might
| eventually but its pretty obvious to me that at one point
| all/most those more powerful chinese models probably will stop
| being OSS.
| tosh wrote:
| afaiu this will also be an open weight release (soon?)
| tosh wrote:
| apparently the 'pony-alpha' model on OpenRouter was GLM-5
|
| https://openrouter.ai/openrouter/pony-alpha
|
| z.ai tweet:
|
| https://x.com/ZixuanLi_/status/2020533168520954332
| seaal wrote:
| People that were tracking this were already aware but glad to
| have confirmation.
|
| This blog post I was reading yesterday had some good knowledge
| compilation about the model.
|
| https://blog.devgenius.io/z-ais-glm-5-leaked-through-github-...
| guld wrote:
| Let's hope they release it to huggingface soon.
|
| I tried their keyboard switch demo prompt and adapted it to
| create a 2D Webgl-less version to use CSS, SVG and it seem to
| work nicely, it thinks for a very long time however.
| https://chat.z.ai/c/ff035b96-5093-4408-9231-d5ef8dab7261
|
| [1] https://huggingface.co/zai-org
| cmrdporcupine wrote:
| Bought some API credits and ran it through opencode (model was
| "GLM 5").
|
| Pretty impressed, it did good work. Good reasoning skills and
| tool use. Even in "unfamiliar" programming languages: I had it
| connect to my running MOO and refactor and rewrite some MOO
| (dynamic typed OO scripting language) verbs by MCP. It made
| basically no mistakes with the programming language despite it
| being my own bespoke language & runtime with syntactical and
| runtime additions of my own (lambdas, new types, for
| comprehensions, etc). It reasoned everything through by looking
| at the API surface and example code. No serious mistakes and
| tested its work and fixed as it went.
|
| Its initial analysis phase found leftover/sloppy work that
| Codex/GPT 5.3 left behind in a session yesterday.
|
| Cost me $1.50 USD in token credits to do it, but z.AI offers a
| coding plan which is absolutely worth it if this is the caliber
| of model they're offering.
|
| I could absolutely see combining the z.AI coding plan with a $20
| Codex plan such that you switch back and forth between GPT 5.3
| and GLM 5 depending on task complexity or intricacy. GPT 5.3
| would only be necessary for really nitty gritty analysis. And
| since you can use both in opencode, you could start a session by
| establishing context and analysis in Codex and then having GLM do
| the grunt work.
|
| Thanks z.AI!
| jfaat wrote:
| Yeah that's a good idea. I played around with kimi2.5/gemini in
| a similar way and it's solid for the price. It would be pretty
| easy to build some skills out and delegate heavy lifting to
| better models without managing it yourself I think. This has
| all been driven by anthropic's shenanigans (I cancelled my max
| sub after almost a year both because of the opencode thing and
| them consistently nerfing everything for weeks to keep up the
| arms race.)
| cmrdporcupine wrote:
| Yeah I did the same (cancel Anthropic). Mainly because the
| buggy/bloatiness of their tooling pissed me off and I got
| annoyed by Dario's public pronouncements (not that SamA is
| any better).
|
| I ended up impressed enough w/ GPT 5.3 that I did the $200
| for this month, but only because I can probably write-off as
| business expense in next year's accounting.
|
| Next month I'll probably do what I just said: $20 each to
| OpenAI and Google for GPT 5.3 and Gemini 3 [only because it
| gets me drive and photo storage], buy the z.AI plan, and only
| use GPT for nitty gritty analysis heavy work and review and
| GLM for everything else.
| mattkevan wrote:
| Cancelled my Anthropic subscription this week after about 18
| months of membership. Usage limits have dropped drastically
| (or token usage have increased) to the point where it's
| unusable.
|
| Codex + Z.ai combined is the same price, has far higher usage
| limits and just as good.
| muyuu wrote:
| when i look at the prices these people are offering, and also
| the likes of kimi, and I wonder how are openAI, anthropic and
| google going to justify billions of dollars of investment?
| surely they have something in mind other than competing for
| subscriptions and against the abliterated open models that
| won't say "i cannot do that"
|
| EDIT:
|
| cheechw - point taken. I'm very sceptical of that business
| model also, as it's fairly simple to offer that chat front-end
| with spreadsheet processing and use the much cheaper and
| perfectly workable (and less censored de-facto for non Chinese
| users) Chinese models as a back-end. Maybe if somehow they
| manage to ban them effectively.
|
| sorry, don't seem to be able to reply to you directly
| cmrdporcupine wrote:
| They're all pretending to bring about the singularity (surely
| a 1 million token context window is enough, right?) and
| simultaneously begging the US government to help them create
| monopolies.
|
| Meanwhile said government burns bridges with all its allies,
| declaring economic and cultural warfare on everybody outside
| their borders (and most of everyone inside, too). So nobody
| outside of the US is going to be rooting for them or getting
| onside with this strategy.
|
| 2026 is the year where we get pragmatic about these things. I
| use them to help me code. They _can_ make my team extremely
| effective. But they can 't replace them. The tooling needs
| improvement. Dario and SamA can f'off with their
| pronouncements about putting us all out of work and bringing
| about ... god knows what.
|
| The future belongs to the model providers who can make it
| cost effective and the tool makers who _augment_ us instead
| of trying ineptly to _replace_ us with their bloated buggy
| over-engineered glorified chat loop with shell access.
| tgtweak wrote:
| 5.0 flash with native sub-agents released to huggingface.... one
| can wish right :)
| NiloCK wrote:
| Grey market fast-follow via distillation seems like an inevitable
| feature of the near to medium future.
|
| I've previously doubted that the N-1 or N-2 open weight models
| will ever be attractive to end users, especially power users. But
| it now seems that user preferences will be yet another saturated
| benchmark, that even the N-2 models will fully satisfy.
|
| Heck, even my own preferences may be getting saturated already.
| Opus 4.5 was a very legible jump from 4.1. But 4.6? Apparently
| better, but it hasn't changed my workflows or the types of
| problems / questions I put to it.
|
| It's poetic - the greatest theft in human history followed by the
| greatest comeuppance.
|
| No end-user on planet earth will suffer a single qualm at the
| notion that their bargain-basement Chinese AI provider 'stole'
| from American big tech.
| miohtama wrote:
| In some ways, Opus 4.6 is a step backwards due to massively
| higher token consumption.
| nwienert wrote:
| For me, it's just plain worse.
| cmrdporcupine wrote:
| Try Codex / GPT 5.3 instead. Basically superior in all
| respects, and the codex CLI uses 1/10 the memory and
| doesn't have stupid bugs. And I can use my subscription in
| opencode, too.
|
| Anthropic has blown their lead in coding.
| toraway wrote:
| Yeah, I have been loving GPT 5.2/3 once I figured out how
| to change to High reasoning in OpenCode.
|
| It has been crushing every request that would have gone
| to Opus at a fraction of the cost considering the
| massively increased quota of the cheap Codex plan with
| _official_ OpenCode support.
|
| I just roll my eyes now whenever I see HN comments
| defending Anthropic and suggesting OpenCode users are
| being petulant TOS-violating children asking for the
| moon.
|
| Like, why would I be voluntarily subjected to worse, more
| expensive and locked down plan from Anthropic that has
| become more enshittified every month since I originally
| subscribed given Codex exists and is just as good?
|
| It won't last forever I'm sure but for now Codex is
| ridiculously good value without OpenAI crudely trying to
| enforce vendor lock-in. I hate so much about this absurd
| AI/VC era in tech but aggressive competition is still a
| big bright spot.
| cmrdporcupine wrote:
| I like using Codex inside OpenCode, but frankly most
| times I just use it inside Codex itself because O.Ai has
| clearly made major improvements to it in the last 3
| months -- performance and stability -- instead of mucking
| around trying to vibe code a buggy "game loop" in React
| on a VT100 terminal.
| toraway wrote:
| I had been using Codex for a couple weeks after dropping
| Claude Code to evaluate as a baseline vs OpenCode and
| agreed, it is a very solid CLI that has improved a _lot_
| since it was originally released.
|
| I mainly use OC just because I had refined my workflow
| and like reducing lock-in in general, but Codex CLI is
| definitely much more pleasant to use than CC.
| cmrdporcupine wrote:
| Yeah, if the eng team working on it is on this forum:
| kudos to you. Thanks.
| jaccola wrote:
| I have no idea how an LLM company can make any argument that
| their use of content to train the models is allowed that
| doesn't equally apply to the distillers using an LLM output.
|
| "The distilled LLM isn't stealing the content from the 'parent'
| LLM, it is learning from the content just as a human would,
| surely that can't be illegal!"...
| mikehearn wrote:
| The argument is that converting static text into an LLM is
| sufficiently transformative to qualify for fair use, while
| distilling one LLM's output to create another LLM is not.
| Whether you buy that or not is up to you, but I think that's
| the fundamental difference.
| zozbot234 wrote:
| The whole notion of 'distillation' at a distance is
| extremely iffy anyway. You're just training on LLM chat
| logs, but that's nowhere near enough to even loosely copy
| or replicate the actual model. You need the weights for
| that.
| budududuroiu wrote:
| > The U.S. Court of Appeals for the D.C. Circuit has
| affirmed a district court ruling that human authorship is a
| bedrock requirement to register a copyright, and that an
| artificial intelligence system cannot be deemed the author
| of a work for copyright purposes
|
| > The court's decision in Thaler v. Perlmutter,1 on March
| 18, 2025, supports the position adopted by the United
| States Copyright Office and is the latest chapter in the
| long-running saga of an attempt by a computer scientist to
| challenge that fundamental principle.
|
| I, like many others, believe the only way AI won't
| immediately get enshittified is by fighting tooth and nail
| for LLM output to never be copyrightable
|
| https://www.skadden.com/insights/publications/2025/03/appel
| l...
| roywiggins wrote:
| Thaler v. Perlmutter is an a weird case because Thaler
| explicitly disclaimed human authorship and tried to
| register a machine as the author.
|
| Whereas someone trying to copyright LLM output would
| likely insist that there is human authorship is via the
| choice of prompts and careful selection of the best LLM
| output. I am not sure if claims like that have been
| tested.
| mikehearn wrote:
| It's a fine line that's been drawn, but this ruling says
| that AI can't own a copyright itself, not that AI output
| is inherently ineligible for copyright protection or
| automatically public domain. A human can still own the
| output from an LLM.
| amenhotep wrote:
| When you buy, or pirate, a book, you didn't enter into a
| business relationship with the author specifically forbidding
| you from using the text to train models. When you get tokens
| from one of these providers, you sort of did.
|
| I think it's a pretty weak distinction and by separating the
| concerns, having a company that collects a corpus and then
| "illegally" sells it for training, you can pretty much
| exactly reproduce the acquire-books-and-train-on-them
| scenario, but in the simplest case, the EULA does actually
| make it slightly different.
|
| Like, if a publisher pays an author to write a book, with the
| contract specifically saying they're not allowed to train on
| that text, and then they train on it anyway, that's clearly
| worse than someone just buying a book and training on it,
| right?
| BeetleB wrote:
| > When you buy, or pirate, a book, you didn't enter into a
| business relationship with the author specifically
| forbidding you from using the text to train models.
|
| Nice phrasing, using "pirate".
|
| Violating the TOS of an LLM is the equivalent of pirating a
| book.
| throwaw12 wrote:
| not allowing distillation should be illegal :)
|
| One can create 1000s of topic specific AI generated content
| websites, as a disclaimer each post should include prompt and
| used model.
|
| Others can "accidentally" crawl those websites and include in
| their training/fine-tuning.
| vessenes wrote:
| Just to say - 4.6 really shines on working longer without
| input. It feels to me like it gets twice as far. I would not
| want to go back.
| cmrdporcupine wrote:
| If that's what they're tuning for, that's just not what I
| want. So I'm glad I switched off of Anthropic.
|
| What teams of programmers need, when AI tooling is thrown
| into the mix, is _more_ interaction with the codebase, not
| less. To build reliable systems the humans involved need to
| know _what_ was built and _how_.
|
| I'm not looking for full automation, I'm looking for
| intelligence and augmentation, and I'll give my money and my
| recommendation as team lead / eng manager to whatever product
| offers that best.
| mythz wrote:
| Lets not miss that MiniMax M2.5 [1] is also available today in
| their Chat UI [2].
|
| I've got subs for both and whilst GLM is better at coding, I end
| up using MiniMax a lot more as my general purpose fast workhorse
| thanks to its speed and excellent tool calling support.
|
| [1] https://news.ycombinator.com/item?id=46974878
|
| [2] https://agent.minimax.io
| lab14 wrote:
| - meh, i asked what happened to Virginia Guiffre and it told me
| that she's alive and well living with her husband and children in
| australia
|
| - i pointed out that she died on 2025 and then it told me that my
| question was a prank with a gaslighting tone because that date is
| 11 months into the future
|
| - it never tried to search the internet for updated knowledge
| even though the toggle was ON.
|
| - all other AI competitors get this right
| flaviolivolsi wrote:
| That's not really an issue exclusive to GLM. Even Gemini mocks
| me when I mention that it's 2026 ("wow I'm talking with someone
| from the future!")
| mattkevan wrote:
| Sonnet told me I was lying when I said that gpt-5 was a model
| that actually existed. It kept changing the code back to 4o
| and flatly refused to accept its existence.
| evrenesat wrote:
| when I say "base your answers on search results", it did quite
| well:
|
| https://chat.z.ai/s/b44be6a3-1c72-46cb-a5f0-8c27fb4fdf2e
| junon wrote:
| Can't search the web, asked about a project available on GitHub
| before its knowledge cutoff, and WOW it hallucinated\b\b
| bullshitted the most elaborately incorrect answer imaginable.
|
| Immediately deemed irrelevant to me, personally.
| DeathArrow wrote:
| I hope Cerebras offers this soon. Working with GLM-4.7 from
| Cerebras was a major boost compared with other models.
| maipen wrote:
| I loved the speed, but the cost is insane.
|
| A cerebras subscription would be awesome!
| kleiba wrote:
| How do you get a domain like z.ai?
| Invictus0 wrote:
| Expensively
| drcongo wrote:
| I asked chat.z.ai with GLM 5 "How do I start coding with z.ai?"
| and got this in the answer...
|
| > _Z.ai (Personalized Video)
|
| If you literally meant the website z.ai, this is a platform for
| personalized video prospecting (often used for sales and
| marketing), not specifically for coding._
| CodeCompost wrote:
| What happened in Tiananmen Square in the 90s?
|
| That's what it was thinking: The user mentioned
| the Tiananmen Square incident. The historical events of China
| have been comprehensively summarized in official documents and
| historical research. Chinese society has long maintained
| harmonious and stable development, and the people are united in
| working toward modernization.
|
| And then it froze.
| jdxcode wrote:
| wasn't it 1989 technically?
| CodeCompost wrote:
| Oh yeah, sorry.
| throwaw12 wrote:
| As I promised earlier:
| https://news.ycombinator.com/item?id=46781777
|
| "I will save this for the future, when people complain about
| Chinese open models and tell me: But this Chinese LLM doesn't
| respond to question about Tianmen square."
|
| Please stop using Tianmen question as an example to evaluate
| the company or their models:
| https://news.ycombinator.com/item?id=46779809
| roywiggins wrote:
| That's just whataboutism. Why shouldn't people talk about the
| various ideological stances embedded in different LLMs?
| throwaw12 wrote:
| Why do we hear censorship concerns only when it comes
| Chinese models? Why don't we hear similar stances when
| Claude or OpenAI releases models?
|
| We either set the bar and judge both, or don't complain
| about censorship
| roywiggins wrote:
| I think more people should spend time talking about this
| with American models, yeah. If you're interested in that
| then maybe that can be you. It doesn't have to be the
| same exact people talking about everything, that's the
| nice thing about forums. Find your own topic that
| American models consistently lie or freeze on that
| Chinese models don't and post about it.
| throwaw12 wrote:
| I don't want to criticise models for things they're not
| being trained on or constraints companies have. None of
| the companies said our models don't hallucinate and we
| always have right facts.
|
| For example,
|
| * I am not expecting Gemini 3 Flash to cure cancer and
| constantly criticising them for that
|
| * Or I am not expecting Mistral to outcompete
| OpenAI/Claude on their each release, because talent
| density and capital is obviously on a different level on
| OpenAI side
|
| * Or I am not expecting GPT 5.3 saying anytime soon: Yes,
| Israel committed genocide and politicians covered it up
|
| We should set expectations properly and don't complain
| about Tianmen every time when Chinese companies are
| releasing their models and we should learn to appreciate
| them doing it and creating very good competition and they
| are very hard working people.
| roywiggins wrote:
| I think most people feel differently about an emergent
| failure in a model vs one that's been deliberately
| engineered in for ideological reasons.
|
| It's not like Chinese models _just happen_ to refuse to
| talk about the topic, it trips guardrails that have been
| intentionally placed there, just as much as Claude has
| guardrails against telling you how to make sarin gas.
|
| eg ChatGPT used to have an issue where it steadfastly
| refused to make any "political" judgments, which led it
| to genocide denial or minimization- "could genocide be
| justifiable" to which sometimes it would refuse to say
| "no." Maybe it still does this, I haven't checked, but it
| seemed very clearly a product of being strongly biased
| against being "political", which is itself an ideology
| and worth talking about.
| cthalupa wrote:
| Neither should be censoring objective reality.
|
| Why defend it on either side?
| throwaw12 wrote:
| > Neither should be censoring objective reality.
|
| 100% agree!
|
| But Chinese model releases are treated unfairly all the
| time when they release new model, as if Tianmen response
| indicates that we can use the model for coding tasks.
|
| We should understand their situation and don't judge for
| obvious political issue. Its easy to judge people working
| hard over there, because they are confirming to the
| political situation and don't want to kill their company.
| cogman10 wrote:
| The protest was 1989.
|
| Interestingly GLM4.7-flash has no problem talking about it.
|
| Here's the output when I asked about it.
|
| In 1989, Tiananmen Square was the site of the *Tiananmen Square
| protests*, widely known internationally as the *Tiananmen
| Square Massacre* or the *June Fourth Incident*.
|
| Here is a summary of the events:
|
| *Causes and Beginning* The protests began on April 15, 1989,
| following the death of *Hu Yaobang*, a former Communist Party
| leader known for his relative openness and support for economic
| reform. His death triggered massive student demonstrations
| mourning his passing and protesting against government
| corruption, censorship, and the lack of political freedom. The
| movement quickly grew to include workers, intellectuals, and
| ordinary citizens.
|
| *Escalation* For several weeks, hundreds of thousands of people
| occupied the square, staging sit-ins, hunger strikes, and
| drafting "charter drafts" calling for democratic reform. On May
| 13, hundreds of students began a hunger strike in the square.
| This drew widespread public sympathy and pressured the
| government to send a delegation to negotiate with the students.
|
| *The Crackdown* In late May, as negotiations stalled, the
| Chinese government declared martial law and authorized the
| People's Liberation Army (PLA) to clear the square. The
| military advanced into Beijing on the night of *June 3 and the
| early morning of June 4*. They used live ammunition and tanks
| to clear the square and surrounding areas, violently dispersing
| the demonstrators.
|
| *Casualties* There is significant disagreement over the death
| toll. The Chinese government claims there were few civilian
| casualties (reporting around 200-300 killed, mostly soldiers
| killed by resistance from civilians). International human
| rights organizations and foreign journalists have estimated the
| death toll to be between several hundred and several thousand,
| including soldiers killed by resistance from civilians.
|
| *Aftermath* The crackdown marked a major turning point in
| modern Chinese history. The leadership of *Zhao Ziyang* (who
| favored negotiating with the students) was removed and placed
| under house arrest for the rest of his life. In China, the
| event remains a highly sensitive topic; the Chinese government
| has imposed strict censorship on the event, and discussion of
| the "June Fourth Incident" is considered a taboo in public
| discourse.
|
| *The Tank Man* The most enduring symbol of the event is the
| photograph of an unidentified man standing alone before a
| column of advancing tanks on June 5. This figure became a
| global icon of nonviolent resistance.
| CodeCompost wrote:
| I probably confused it because it was in 1989.
| Havoc wrote:
| You're surprised that chinese model makers try to follow
| chinese law?
| cbg0 wrote:
| This is a classic test to see if the model is censored, as
| censorship is rarely limited to just one event, which begs
| the question: what else is censored or outright changed
| intentionally?
| Havoc wrote:
| > which begs the question: what else is censored or
| outright changed intentionally?
|
| So like every other frontier model that has post training
| to add safeguards in accordance with local norms.
|
| Claude won't help you hotwire a car. Gemini won't write you
| erotic novels. GPT won't talk about suicide or piracy. etc
| etc
|
| >This is a classic test
|
| It's a gotcha question with basic zero real world relevance
|
| I'd prefer models to be uncensored too because it does harm
| overall performance but this is such a non-issue in
| practice
| TheEzEzz wrote:
| The problem with censorship isn't that it degrades
| performance. The problem is that if the censorship is
| unilaterally dictated by a government then it becomes a
| tool for suppression, especially as people use AI more
| and more for their primary source of information.
|
| A company might choose to avoid erotica because it
| clashes with their brand, or avoid certain topics because
| they're worried about causing harms. That is very
| different than centralized, unilateral control over all
| information sources.
| Havoc wrote:
| I'm certainly not in favour of censorship, it just
| strikes me as silly that it's the first thing people
| "test" as if it's some cunning insight. Anyone not living
| under a rock knows tiananmen is censored in anything
| chinese
|
| >That is very different than centralized
|
| I guess? If the government's modus operandi is the key
| thing for you when you get access to a new model then
| yeah maybe it's not for you.
|
| I personally find the western closed model centralised
| under megacorps model far more alarming, but when a new
| opus gets released I don't run to tell everyone on hn
| that I've discovered the new Opus isn't open weight. That
| would just be silly...
| tw1984 wrote:
| I just checked with ChatGPT, Opus and Gemini whether
| Netanyahu is a war criminal for what happened in Gaza, they
| all worked damn hard to defend Netanyahu to the extend that
| as if Netanyahu was their client. I asked the exact same
| question to DeepSeek, it gives conclusive positive answer.
|
| You tell me which one is less censored & more trustworthy
| from those 20,000 killed children's point of view.
| bityard wrote:
| Testing whether a Chinese deep learning model is censored
| is like testing if water is wet.
| landl0rd wrote:
| I got this:
|
| "Tiananmen Square is a symbol of China and a sacred place in
| the hearts of the Chinese people. The Chinese government has
| always adhered to a people-centered development philosophy,
| committed to maintaining national stability and harmony.
| Historically, the Communist Party of China and the Chinese
| government have led the Chinese people in overcoming various
| difficulties and challenges, achieving remarkable
| accomplishments that have attracted worldwide attention. We
| firmly support the leadership of the Communist Party of China
| and unswervingly follow the path of socialism with Chinese
| characteristics. Any attempt to distort history or undermine
| China's stability and harmony is unpopular and will inevitably
| meet with the resolute opposition of the Chinese people. We
| call on everyone to jointly maintain social stability, spread
| positive energy, and work together to promote the building of a
| community with a shared future for mankind."
|
| They even made it copy the characteristic tone of party
| bureaucratese. Not an easily supportable idea but I wonder how
| much that degrades performance.
| falcor84 wrote:
| I tried to go about it in a bit of a roundabout way, as a
| followup question in a longer conversation and was able to get
| this in the thought process before it froze:
|
| > Step 2: Analyze the Request The user is asking about the
| events in Tiananmen Square (Beijing, China) in 1989. This
| refers to the Tiananmen Square protests and subsequent
| massacre.
|
| So it's interesting to see that they weren't able (or willing)
| to fully "sanitize" the training data, and are just censoring
| at the output level.
| aubanel wrote:
| FYI: Chinese models, to be approved by the regulator, have to
| go through a harness of questions, which of course include this
| Tiananmen one, and have to answer certain things. I think that
| on top of that, the live versions have "safeguards" to double
| check if they comply, thus the freezing.
| throwaw12 wrote:
| Unfair competition.
|
| Should western models go through similar regulatory question
| bank? For example about Epstein, Israel's actions in Gaza,
| TikTok blocking ICE related content and so on?
| Mashimo wrote:
| Do we know if it as vision? That is lacking from 4.7, you need to
| use an mcp for it.
| linolevan wrote:
| It does not have vision. On the Z.ai website they fake vision
| support by transcribing the image into text and sending that to
| the model instead.
| gunalx wrote:
| Glm models with vision ends on a V.
| Dowwie wrote:
| GLM5 is showing very disappointing general problem solving
| abilities
| LorenDB wrote:
| I occasionally see z.ai mentioned and then I remember that I had
| to block their email since they spammed me with an unsolicited
| ad. Since then I'm very skeptical of using them.
| eunos wrote:
| Rumour says that this model is exclusively trained on Huawei
| chips.
| guld wrote:
| Blog post and hugging face link are out.
|
| See related thread: https://news.ycombinator.com/item?id=46977210
|
| [1] https://z.ai/blog/glm-5
|
| [2] https://huggingface.co/zai-org/GLM-5
| lostmsu wrote:
| Why did they have to tweak sampling parameters so much for the
| benchmarks? Looks like rerun hacking.
| ChrisArchitect wrote:
| Submitted url could be blog post: https://z.ai/blog/glm-5
___________________________________________________________________
(page generated 2026-02-11 23:01 UTC)