Asking very specific questions to Sonnet 3.7 costs a couple of tenths of a cent every time, and even if you're doing that all day it will never amount to more than maybe a dollar at the end of the day.

On average, one line of, say, JavaScript represents around 7 tokens, which means there are around 140k lines of JS per million tokens.

On Openrouter, Sonnet 3.7 costs are currently:

- $3 / one million input tokens => $100 = 33.3 million input tokens = 420k lines of JS code

- $15 / one million output tokens => $100 = 3.6 million output tokens = 4.6 million lines of JS code

For one developer? In one day? It seems that one can only reach such amounts if the whole codebase is sent again as context with each and every interaction (maybe even with every keystroke for type completion?) -- and that seems incredibly wasteful?

I can't edit the above comment, but there's obviously an error in the math! ;-) Doesn't change the point I was trying to make, but putting this here for the record.

33.3 million input tokens / 7 tokens per loc = 4.8 million locs

3.6 million output tokens / 7 tokens per loc = 515k locs

That's how it works, everything is recomputed again every additional prompt. But it can cache the state of things and restore for a lower fee, and reingesting what was formerly output is cheaper than making new output (serial bottleneck) so sometimes there is a discount there.

I'm waiting for the day this AI bubble bursts since as far as we can tell almost all these AI "providers" are operating at a loss. I wonder if this billing model actually makes profit or if it's still just burning cash in hopes of AGI being around the corner. We have yet to see a product that is useful and affordable enough to justify the cost.

It's burning cash. Lots of it.

[0] https://www.wheresyoured.at/openai-is-a-systemic-risk-to-the...

Great article, thanks. Mirrors exactly what the JP Morgan/Goldman report claimed but that was quite dated.

It sounds insane until you drive full agentic loops/evals. I'm currently making a self-compiling compiler; no doubt you'll hear/see about it soon. The other night, I fell asleep and woke up with interface dynamic dispatch using vtables with runtime type information and generic interface support implemented...

Do you actually understand the code Claude wrote?

Do you understand all of the code in the libraries that your applications depend on? Or your coworker for that matter?

All of the gate keeping around llm code tools are amusing. But whatever, I’m shipping 10x and making money doing it.

Up until recently I could be sure they were written by a human.

But if you are making money by using LLMs to write code then all power to you. I just despair at the idea of trillions of lines of LLM generated code.

Well, you can’t just vibe code something useful into existence despite all the marketing. You have to be very intentional about which libraries it can use, code style etc. Make sure it has the proper specifications and context. And review the code, of course.

Fair enough. That's pretty cool, I haven't gone that far in my own work with AI yet, but now I am inspired to try.

The point is to get a pipeline working, cost can be optimized down after.