Comment by binary0010
1 day ago
So how do openai and anthropic plan to keep customers when GLM-5.1 is just as good and open source and a lot cheaper?
I don't see the business model working. My closest friend actually does automation software for large companies.
He does not use Claude or openai at all. He primarily uses gpt 120b on cerebras and glm-5.1 for heavy thinking work. And some other small models for various tasks. All open source.
And these systems are extremely useful for the businesses and are able to run fully automated pipelines that are very stable and fast.
We discuss this a lot, and we both think any business doing heavy agentic work on Claude and openai just aren't aware of exactly how good and cheap open source has gotten on the last year.
So... once the legacy businesses and developers catch up, won't Claude and openai be unable to recoup their costs?
> I don't see the business model working.
Same. It's a nightmare from a Porter's Five Forces perspective.
There will be a ton of businesses competing in this space, and there will be something of a moat due to how capital intensive the business can be, but there will still basically be infinite competitors.
Great for consumers.
Well in reality AWS will just host one of them and most companies will use that
Like how snapchat kind of fell off because the feature could just be a subset of instagram
It seems like it would just become a commodity like EC2
Snapchat is huge and growing
2 replies →
For coding you always want to go with the best model in the category, not something that would be the best model if we went 1 year back which GLM 5.1 is, and I'm saying that as a big fan of GLM cause I run a translation site where GLM is good enough for the price.
Most of the money right now is in coding. Openai and Anthropic just have to be 6 months ahead of SOTA open source models and they'll capture most of the enterprise and dev market
Yes I'm an engineer (20 years most in games/graphics industry) and only use it for code. I've been using glm 5.1 this week a lot. I went in expecting another "decent" but not really "up to standard" open source model.
I highly doubt I'll ever use Claude again.
I think you are wrong about Claude being any significant level better
I've been mostly coding with GLM-5.1 as well and I agree with you. DeepSeek V4 Flash is another very good surprise. Incredibly cheap, fast and effective.
1 reply →
Well I think there are a multitude of harder measurements that would disagree with you, but ultimately there is absolutely a use case for cheaper open models (or even cheaper tiers of proprietary models) and in fact the unsolved optimization everyone is trying to get to is how much spend to use for a given task. But there will always be a market, especially in enterprise, for the best performance there is to offer
2 replies →
I strongly disagree. I'm an engineer - I'm all about the fastest, cheapest thing that meets the requirements. I don't need Opus 4.7, even for my complex programming tasks. It costs over 10x other models available that still give good enough answers. Those smaller models are also a lot faster to output tokens, which saves me time.
Once the model gets good enough, the returns on bigger models diminishes quickly. I don't want to spend 10x the money and wait 5x the time to get answers that are equivalent.
Same here, i can't say i've seen any difference in 4.6 vs 4.7 other than price
If I generate code with Claude, ChatGPT, and GLM 5.1, I can't say which model is which reliably. I exclusively use Claude more out of superstition than reason.
> For coding you always want to go with the best model in the category
This is transparently false, because the best "model" is still competent human developers. They're just more expensive. If you're willing to use current LLMs at all, it means you're willing to sacrifice quality for a better price, and your disagreement with the comment you were replying to is entirely about what the optimum tradeoff is.
Well it may be false that you always want the best model, but the point is performance of you+<agent> is far more cost effective than you+someone else
2 replies →
It was true 6 months ago, not anymore. Frontier models now outperform developers on many tasks, be it on quality/readability/maintainability, and let’s not talk about speed…
13 replies →
I have stats from a harness that tells me glm5.1 is far more cost effective for us than Opus with the rate of defects and rework taken into account. In fact, with a decent harness I'm now increasingly favouring eHaiku over Opus for execution too. Opus is still worth it for planning, though, and far better at one-shotting things.
> For coding you always want to go with the best model in the category [..]
And this is why many companies go out of business. You always want the best bang for your buck, sometimes this is the "best model" and sometimes it is not.
For coding like for everything else in life cost is a factor.
Cost for the value delivered. Like if you offered the current SOTA open source models at $0.1/M, I still think I'd be using Opus or 5.5 at $30/M. Or say GPT 5 which was released Aug 25, I don't think I'd use it for coding for even $0.1. I'd def find other uses for it(translations, agentic workflows, prompt guards etc), but for coding I don't think I'd ever completely switch to a SOTA open model
Unless ofc there was an actual speed difference, only reason I'd be willing to go with a worse model couple of percent worse than current best model is if the speed was at least 5x higher. Looking forward to kimi k2.6 offered publicly by Cerebras
4 replies →
Why? If it's good enough, it's good enough. Though I read the code that gets vibed so maybe my use-case is different.
It's driven a lot by the harness too. If you're using claude code, you're actively being pushed towards newer models, even though older ones work perfectly fine for your use cases
>For coding you always want to go with the best model in the category, not something that would be the best model if we went 1 year back which GLM 5.1 is, and I'm saying that as a big fan of GLM cause I run a translation site where GLM is good enough for the price.
Currently, the difference is substantial, but what happens if capabilities saturate?
Then the house of cards comes crumbling down, but there is so much evidence to point to this not happening that it requires a bit of a theory for how that may happen
2 replies →
And you propose the same companies that have been cost cutting and avoiding buying you a chair for ever won't start objecting to a $200/dev/month subscription? The finance department won't have a say?
> For coding you always want to go with the best model in the category
Will this always be true? There will never be an event horizon/point of diminishing returns where something not-bleeding-edge is "good enough" for 51%+ of users?
As long as closed source is 6 months ahead in terms of current difference. Although this is hard to figure out using simple percent based coding benchmarks, you def. notice it when you're actually trying to do a long task. Even simple things like UI "taste" is enough for me to use opus instead of 5.5 though even though 5.5 is strictly better for anything that doesn't have a UI, ie backend, scripts, making agent workflows etc
This is a silly take. There is a line of "good enough" for most coding (most CRUD apps and APIs are nothing special), and once we are past that, nobody will care about having the "newest, best" model except extreme outliers. And this base "good enough" model will become an ultra cheap commodity as we already see with GLM, deepseek, etc.
As long as closed models are 6 months ahead I won't be switching from them to prev. 6 month SOTA open source models. Maybe its just a different calculation if you're in a job, but as an indiehacker I'll take any edge I can get
Ofc again, can be convinced to switch if there's however a clear speed difference, like 5x+ for a open source sota even if it was SOTA for 6 months ago
> For XXX you always want to go with XXX, not XXX
Oh, hey, I recognize you. Thank you for the very forward and thorough orbital sander recommendation at Home Depot. That's exactly what I wanted to deal with on my holiday weekend. You just know so much about this and the rest of us are simple passersbys.
Yep sorry was just pulling it out my rear, not like a market trend that nearly every enterprise uses Anthropic or Openai models for coding or that Anthropic has had such ridiculous growth that they're 10x-ing year over year
1 reply →
Most work is not coding.
And also, people have it wrong… their models are not the main problem anymore. It’s the RAG
Would love to hear more about your thought about the RAG.
1 reply →
Depending on RAG is a workflow problem, not an AI problem
For coding assistance, I have tried OpenCode with several large open models through OpenRouter. All were fairly bad compared to Claude Opus. Could you provide some hints on how I should be holding these open models so that I might get more value out of them?
I agree with the common trope that open models lag behind by about a year, but something magical happened just around a year ago when the state of the art models became extremely useful. By this reasoning we're about to see open models perform well, but I'm afraid there is more to it than just waiting for another revolution around the sun.
Note, my application is coding assistance. Open models can be great for other purposes.
I tried almost all OS models on opencode, none of them is on levels as opus 4.7.
In latest experiment I used opus for implementation plan then used cursor composer 2.5 for execution.
I must say that combo is really good. Main drawback of claude code is that is super slow. So when paired with composer that is super fast it flies.
No one is claiming that OS is as good. They are saying it isn't that far behind SOTA commercial products. So why pay exorbitantly just to get something only a few percent better than the free option?
But there have been very good open source office apps for decades and few enterprises use them, so perhaps this is just the nature of B2B purchasing committees and 'nobody getting fired for buying IBM.'
1 reply →
Do more planning yourself, be smart about the context, break down tasks into smaller components, give it more guidance. You can't just lazily prompt it to complete large features autonomously and expect good results.
But if the closed-source models can do this without the additional effort, that's a significant gap, no?
7 replies →
+1 .. just wanted to reiterate that this is the answer. The open models work great if you just do a little more of the design/architectural work up front and organize your work appropriately.
a good harness is supposed to do what you are describing. sonnet on pi.dev is pretty terrible but fast. Claude Code has ridiculous amounts of prompt engineering at system prompt level and sub session spawing combined with low temperature, to provide the predictable results people like. CC screws up and you never see, because the harness auto corrects, while on OSS you see everything, and does not comes with the level of monitoring by default.
GLM-5.1 isn't just as good. It is no match for Opus running in Claude Code. Please try it yourself. Open source models are about a year behind at least.
This is profoundly misinformed. I use all three of those models regularly and the difference is just not that big anymore. GLM 5.1 is at least as good as Opus 4.5 - when it’s my dime it’s the primary model I use and switch to GPT 5.5 for planning and review but it’s also very capable at those things. If I had to pay API rates for everything there is no question I would only use GLM 5.1 (and Minimax for exploration tasks).
At work I mostly use Claude Code and a bit of Codex; personal projects are OpenCode and honestly I prefer it.
I would agree here. And in my experience Qwen 27B and Deepseek v4 are also extremely good.
None of them are quite opus, but they are damned close and a no brainer if you care at all about cost.
In the second half of last year, I found that agentic coding with proprietary models (≈ vibe coding) reached the point where it actually speeds up my ability to deliver useful code at work. Before that, AI-based autocomplete definitely helped, but (despite the claims of the people selling AI coding tools) letting an agent author more than a file or so at a time (often a function or so at a time) required a very intricate plan or it would create a mess. Creating that plan or cleaning up the mess would take longer than just doing everything myself.
For me, it feels like widely available open models have recently crossed that same canyon. Are they as good as e.g. late-model Claude Opus? I don't think so. But they have absolutely gotten past the point where they are beneficial. This means that, for me, they are about six months behind.
Exactly this. GLM 5.1 is the first open model that I thought "actually worked" for agentic coding, which puts it in the same tier as Opus 4.5 - which was where I flipped.
For coding I wouldn't say a year, last year this time claude or gpt definitely weren't able to do what GLM is able to do today, but easily 6 months I'd say.
Not sure about other domains though.
A year behind is still very very very good at this price. ;)
I use composer-2 daily for complex programming tasks. It's a fine tuned Kimi 2.5 - nothing groundbreaking. I've even had reasonable success using Qwen 3.5 on my desktop GPU. Opus might be better, but it's certainly not necessary to get good results.
I have 26 years experience. I code using GLM-5.1. Fron time to time I switch to Codex / Claude, and honestly I don't understand why people uses Claude or codex. With the right prompting, GLM is awesome.
Don't you need to spend 5-10 thousand USD to run these models that are "as good" as frontier models from 6-12 months ago? I haven't seen a convincing breakdown for ROI of running your own coding models. Especially against a $20 or even $200 plan
I assume you can run them in the cloud. $5-10k doesn't sound like remotely enough to run a not-shit model locally based on my experience.
Agree. Also reasonix with deepseek is super cheap and quality is only slightly worse (in my experience)
Do you have a good source to refer to, to map out migration from Claude code to a cheap setup using small open source models like you’re describing? I’d certainly like to experience how good they’ve gotten.
The only way I see it working out for them is if some legislation is passed that eliminates the competition by making it illegal to run local models. They could claim that the models are dangerous and could be weaponized without oversight, or something along those lines.
They are both (and also spacex) sprinting for IPOs. They know that the opportunity window is closing fast and that advancement in model quality has largely plateaued in the last year. Take as much investor money as you can get away with for now.