Hacker News Reader

> For one, on the margin people are willing to pay a lot for slightly better models. I know personally the value the LLM adds to my workflow is considerably more than the $200/m I pay the frontier labs. I have no interest in optimizing that to get it slightly lower. There are a very vocal minority that optimizes this or companies whose LLM expense is marginal, but I think that's the minority (correct me if I'm wrong, curious what their customer base looks like)

Enterprise is how Claude/OAI make their money. Enterprise companies absolutely optimize around costs for infra, it's like an entire job category.

> Also the actual LLM is a tiny portion of the value added. Anyone that tried to build agentic solutions from LLM apis quickly realizes that a huge value is the Claude Code / Codex harness. There are open source implementations like OpenCode but they're not nearly as good.

Claude and OAI are not valued at $1T because of their harnesses

> Think about it another way. Consider how much money Microsoft spends on maintaining Excel. There are open source alternatives that have >90% of the functionality, they'll even work w/ Excel files and generate them. Google sheets is probably 99% and available to everyone and better in a lot of regards. But the immense value spreadsheet software produces workers above the $100 or whatever a year makes it so that there is a real moat and no one bothers exploring alternatives.

Enterprise excel is like $50 a month, the closed source labs charge orders of magnitude more than that per user per month for enterprise, and want to charge more if they can to maintain their valuation. I spent $20 in an afternoon yesterday using my work plan for one workflow for reference, I probably spend at least $50 a day and I don't even do that much coding anymore.

$200/mo for 20x Max is great value - but it's not sustainable and won't last forever. Microsoft gave up. Anthropic will too. I use roughly $10k/mo - at that price it's absolutely not worth it.

I'm genuinely not sure where the balancing point even is.

Probably increasing learning rate have some value for these players which is enabled by subsidized usage.

And cost of acquiring learning via users feedback versus additional learning data value for a given company will dictate whether such company will give up or will continue subsidized usage, despite being unprofitable on the paper, but probably really valuable as a long term strategy for GTM and product development and increased learning rates.

> Claude and OAI are not valued at $1T because of their harnesses

They're valued at that because they add a lot of value and people pay for the product. The product is more than the LLM. If you want argue the value of the harness vs LLM but flippant remark adds nothing.

> Enterprise excel is like $50 a month, the closed source labs charge orders of magnitude more than that per user per month for enterprise, and want to charge more if they can to maintain their valuation. I spent $20 in an afternoon yesterday using my work plan for one workflow for reference, I probably spend at least $50 a day and I don't even do that much coding anymore.

I want to know what percent of their revenue is this API spend compared to just a Claude/Codex max plan that's $200/m and very generous.

I'd imagine it has to be immense. Most of the engineers I know are spending $200/day, let's say 20 workdays per month, $4000/mo. Whatever % of revenue it is, it's got to be close to 100% of profit.

Who is paying $200/day? My company has me on a $30/month plan (I assume they pay annually?). I use Claude constantly and only hit usage limits when I try to do 4+ projects at once.

I can't imagine why anyone would need almost 7x that.

If your company is a certain size or cares about ZDR, you need to be on an Enterprise plan which is API pricing. $300/day here on average.

Depending on the harness, $200/session can be common. Thanks for harnesses like maki the cost is lower compared to OpenCode or Cursor.

But $200/day is very common in the business. The monthly plans are just a trial version of what's going to come for all of us. That's why I am very carefully looking into the open models today.

Here, people tend to forget about enterprise customers. Enterprise is excluded from using these heavily subsidized subscriptions, I know of orgs that are spending around $500k/month for teams of ~100 developers actively using AI.

This is a spot where smaller orgs are getting a definite advantage (both cost and speed). If you're a company with 20 - 50 people, you can probably get away with Claude Teams instead of Claude Enterprise, and pay 10x less in ai costs.

> considerably more than the $200/m I pay the frontier labs.

This is a very rich / developed country privilege perspective.

Where I live, it's not unusual for a monthly wage to be around $200. Of course developer wages are much higher, maybe as much as $1000 a month, but $200 is still a huge chunk of that so it doesn't really matter how much "value" you get out of it if you're no longer able to pay rent or buy decent food.

Even in developed countries, $200 a month is out of reach for all kinds of people who would benefit from it (students without rich families, entrepreneurs, etc.).

Thats great, but the market will always be driven by rich developed countries. I don't think Anthropic and OpenAI is banking on third world adoption in their growth estimates.

> a huge value is the […] Codex harness. There are open source implementations

Like Codex https://github.com/openai/codex

If you think Codex is some neutral harness designed to work equally as well for all LLMs, I got a bridge to sell you...

> on the margin people are willing to pay a lot for slightly better models

That margin is getting smaller and smaller. I would have been with you a week ago; paying for Fable was worth it compared to all other models. But with K3, the difference has shrunk to the point where, for me, it's not worth the cost anymore.

In other words, it may be worth paying five times as much to get 10% better real-world outcomes for a lot of people, but a lot fewer people will pay five times as much to get 2% better outcomes.

> a huge value is the Claude Code / Codex harness

For me, it's the opposite. Having to use Claude Code instead of the harness I prefer is a point against Anthropic, not for it.

100% agree for the harnesses, I always end up using CC when I want to use Opus but a few months ago I had my opencode harness set up but Copilot Pro+ rose too much the prices for me to keep using it… What harness do you prefer using ?

OpenCode because of the large ecosystem. I'd like to use pi, but it's more limited if I want to run it in a VM and connect to it from the outside.

I don't think subs are what keep their lights on. And their API is so absurdly expensive. For the harness hard disagree but ultimately its up to each ones taste. You may want to check this tho https://harnessrank.net

On our side we use Claude/GPT/Kimi (it replaced Antigravity) for development. But we build our systems around a cheaper denominator (Deepseek previously, recently we added GPT 5.6 which have good prices as well). We offer BYOK for Claude but its def not an option to build something on top of it (for us).

> There are open source alternatives that have >90% of the functionality, they'll even work w/ Excel files and generate them

Excel has network effects. If N people are using excel, the incremental N+1'th user is forced to use it or risk not being able to open documents from the N users. Excel is also sticky, in the sense that a user used to Excel UI is hesitant to switch to a new UI. Tokens dont have network effects. If N people use fable, the N+1'th user could use any other model, and have no impact.

Not true. Enterprise pay for api. If i am allowed to use fable withouta budget, i can spend 3k per day. Using combination of cheaper models (and better at least in my experience) can lower this to 300. Considering the layoffs, thw companiesdo care a lot of this cost. Price matters more than quality for many use cases. For example, i can writea game where npc are all k3/5.6 level intelligence if the price is cheap enough.

The $200 you’re paying is heavily subsidized, you need to consider the real inference cost which could be closer to $10k if you’re using the maximum permitted by your plan.

The "real" cost is what people pay, not what the AI companies claim.

May I know the source for "closer to $10k"?

> I know personally the value the LLM adds to my workflow is considerably more than the $200/m I pay the frontier labs.

AFAIK the frontier labs make their money on enterprise, and they recently changed their terms so that most businesses can't use the $200/m plan anymore.

This is why companies like Uber - and mine :( - announced per employee AI budgets on the order of ~$1000/month.

When you're forced to pay API costs, the value prop of Kimi/Qwen becomes a lot more compelling.

I'm not sure about the harness now: https://github.com/lidge-jun/opencodex

You assume everyone has extra 200$ to pay. For someone living in a low wage country, it's hard to pay that much, even if you produce a multiple of that value in theory

Spending a lot on making your top developers more effective is easy to justify.

But once you have an AI-powered application in production, why wouldn't you go with far cheaper and capable enough models? That's just optimising a business process like any other. You'd use cheaper providers whether Chinese, or other on-prem models.

What is your workflow such that frontier models increases productivity considerably?

For people cranking out SaaS / web services / web pages / "full stack" work I don't see a huge difference.

If you're building an optimizing compiler, a CUDA kernel, a database, a high performance concurrent data structure with tricky locking, etc. etc. it's still not really close. Sol 5.6 on high just slays e.g. GLM 5.2 for this kind of work for me.

I'm sure K3 is fine for these things too, but I can't afford it at its API rates compared to a Codex coding plan.

For now.

Regarding the latter, are you able to obtain good results from an LLM? With graphics programming work, I find LLMs only help in cases where the task would take me 5 mins or so.

Absolutely.

Interesting. Do you happen to have a public example?

Of what specifically?

A CUDA kernel, compiler optimization, or anything of similar complexity

Feel free to look, but don't bother replying if you have something negative to say, don't feel like having my day ruined by negativity.

https://github.com/rdaum/mica/tree/main/crates/relation-kern...

https://github.com/rdaum/mica/tree/main/crates/vm

https://github.com/rdaum/pagebox/blob/main/crates/wal/src/wa...

Thanks for sharing. Looks interesting. Was all the code AI generated, or is it a mix?

that stuff in particular -- AI generated with heavy heavy prompting and up front design work and post-implementation testing

I have CUDA work here somewhere too but I have the repository private right now

Very cool. Thanks for sharing.

Here's what is key: never rely fully on the "intelligence" of the model. Build a workflow and/or tooling that allows for empirical improvement. e.g. for performance tuning I have a benchmarking framework and a container/server that contains the differential results available via MCP for the agent to observe as it works. Using /goal and a clear destination you want to get to, it will literally grind for hours and use that help. Doesn't help with architectural stuff, but it does help with producing efficient code.

I’m on the $200 claude plan, and keep getting booted down to opus or even sonnet by the broken guardrails.

Why should I pay for a frontier model that I can’t use?

This is a bad take.

> Think about it another way. Consider how much money Microsoft spends on maintaining Excel. There are open source alternatives that have >90% of the functionality, they'll even work w/ Excel files and generate them.

Sure but were these open source alternative as pervasive and widespread and talked about when Microsoft was bundling Excel with the OS? Or were there millions of dollars, if not billions of dollars being spent by these competitors? The answer is No. It is not the same scenario.

While I understand the logic of harnesses - those are not full proof. It is trivial to setup K3 or Qwen to work with Claude Code/Codex by intercepting requests and routing to K3/Qwen. You have tools like CCSwitch which can do that for you.

The only thing remains is - people willing to pay a lot for slightly better model. That is true but price sensitivity is also a thing. The mania case for Anthropic/OpenAI/SpaceX/Google is that they will capture large part of the enterprise market. But in many cases outside coding AI capabilities will be resold.

That is instead of buying AI for lets say HR functions you pay for embedded AI in your HR software. And the HR software company will have incentives to raise their own margins. If their provide similar looking experience using Qwen/K3 why will they buy Anthropic/OpenAI. That is the real risk here. While as an individual you can continue to pay top dollar.