> Spent over $400,000 in API priced tokens with GPT-5.6 Sol and GPT 6 Astra
Tangent here, but I think this bit is super interesting!
You could viably hire someone to do this work for that kind of money - I think the interesting thing is that substantially less interested/experimenting engineers would consider paying for a human to do this work, than would happily chuck a big amount of money into an LLM.
I don't have any suggestion about why that exists, but it's a strange and interesting contract.
This is Theo's project, and he has been very transparent about how he is only doing these things because api prices are heavily subsidized by subscription pricing. The reason people aren't willing to pay 400k to an engineer to do this work is because they aren't paying the AI this much.
I don't think I've seen an enterprise paying API prices doing these sorts of rewrites, and honestly, I think until that happens I will remain skeptical about LLM AI's viability as a profitable business.
I found the whole section super interesting... I'll copy it here:
I used a lot of OpenAI models to try and complete this port. In total I did over $400,000 in API priced tokens with GPT-5.6 Sol and GPT 6 Astra. They wrote over 1.3m lines of Rust over multiple months of /goal loops and never got past like 84% compat.
When I saw how little my Claude Code limits were burning, I figured it'd be fun to throw Opus 5.5 at this. It had a working v0 in 10 hours.
I assumed it kept using the code the Codex models wrote. I was wrong. Opus 5.5 started from scratch. It got further than Astra in 1/10th the time.
I let it keep going, and it definitely did. Total token spend was ~$24,047 of API spend over 2 weeks. I was using my Claude accounts, and it worked out to somewhere between 925% and 983% of my $200 plan weekly limits.
Expensive, for sure, but not that bad considering how much work has went into typescript-go.
>Spent over $400,000 in API priced tokens with GPT-5.6 Sol and GPT 6 Astra. They wrote over 1.3m lines of Rust over multiple months of /goal loops and never got past like 84% compat.
Then they let Claude code lose on the problem:
> I figured it'd be fun to throw Opus 5.5 at this. It had a working v0 in 10 hours. I assumed it kept using the code the Codex models wrote. I was wrong. Opus 5.5 started from scratch. It got further than Astra in 1/10th the time.
In the warnings from README.md
> Also worth mentioning: I've never read a line of this code.
TypeScript team member here - this is impressive work, and it's honestly crazy that 3 of these ports have popped up in the last week! I'm currently AFK but we're hoping to learn more and we'll have more to say on this soon.
I suppose my comment sort of implies this. I think the naive knee jerk reaction is that if you can spit out perfect machine level code against a specification, then logically being closer to bare metal would help with performance. However, without knowing the details of how things worked it is tough to say whether it is truly "reasonable" or not.
But more importantly, the Typescript team themselves [1] picked Golang mostly out of ergonomics and ease of porting at the time. It is becoming more increasingly more to "taste" what ergonomics means. Notably, they did not pick it because they believed it to be the fastest option. But speed was on the mind, giving way to ergonomics and ease of porting. At least, this is my read.
Language and compiler design/ implementation is very much a human task. There's lots of nuances to consider. I doubt they want agent spaghetti code in their codebase either.
GC doesn’t affect performance (at least visibly), memory allocations on critical paths do. Moreover, Rust ARC may cause memory fragmentation and perform worse than a GC.
Writing Rust, Zig, or Go, you would still control memory manually, where it matters.
It does have an impact. Discord blogged about unavoidable GC delays in go that occurred because go needed to traverse live memory, their only option (in go) for reducing that pause was to reduce the size of their data set.
It depends on the workload. ARC can be more expensive than tracing GC, especially with heavily shared objects.
But my main point was that the presence of a GC has nothing to do with how close a language is to the metal. Memory management strategy and low-level capabilities are two separate things.
I’m not saying it doesn't work. I’m saying it’s a waste of resources and won’t gain widespread popularity. The current trend is porting source code to languages that generate native code.
I dislike the WebAssembly advocacy of some folks, where they act as if there weren't similar attempts all the way back to UNCOL in 1958, and even in recent history, seem to only focus on JVM, while ignoring the CLR and BEAM.
However, there are indeed some cases where it does make sense since we have it around anyway, and one of them is running the Typescript compiler in the browser, now that it is written in a compiled language.
Theo is making a new twist on the idea of IDE which hosts the application itself. I'm sure there's going to be a video about it soon. But the idea is to embed the whole compilation and type-check in wasm in the browser which then runs the app immediately, as the agents edit the code.
Wasm is a first-class target for Rust, which produces faster and smaller Wasm binaries. My understanding is that you need to use TinyGo to reasonably target Wasm with Go.
Go having a garbage collector isn't necessarily bad, since Wasm has GC, but I think Go's GC can't easily use Wasm GC because Go has interior pointers.
I'm assuming they mean because it doesn't need a garbage collector or much of a runtime at all. You can target direct to WASM and use I guess WASI, while Go brings with it a garbage collector and so on.
Was it ever a reasonable choice? At the time the choice was made the tooling ecosystem was already split between Rust and JS, and by putting TS in Go they chose to split it three ways. Why not split it 4 or 5 ways then? The pressure will always be towards less duplicated work.
The only real choice of language to build the next generation of JS tools in is JS. Anything else is a vote of no confidence in ourselves.
It's ok for a language (JS in this instance) to be good in a domain but not good in all domains. Pulling a language in every direction forces it to make compromises that hurt its applicability in specific areas.
I think the future of this is -- get your AI agent to investigate this port for anything actually worthwhile, comparing it to the go version, and then take those good ideas and actually maybe bring them into the real compiler.
What a waste of power... $424.000 of API priced tokens amounts to how many kWh burned? And how would you trust this to build your code?
Motivations says it all:
Motivations
Test model capabilities
Make a fast TypeScript type checker
Make a ts checker that can work in WASM with high performance
Memes
Maybe there are just so many of these but at this point for vibe coded stuff like this I’m just like “who cares.” LLMs can make stuff like this now. Unless this reaches some critical mass of usage among developers why would I use it? It’s just a less maintained, less tested implementation. The more important question is “what does this teach us?” And idk the answer to that one.
Yes. This has nothing to do with people trying to rewrite things in Rust for fun. It's a part of his development project that I'm sure will get a video in the future.
I think the difference here is that the original motivation for choosing Go over Rust was that Go was easier to port TypeScript to.
LLMs change the calculus a lot, the result seems faster, and it's possible that the TypeScript team might want to change directions. It'd be a big deal, but I bet they'll at least discuss it.
Generally building something is not about "What does this teach us?" but rather "What can this be used for?". You don't build a house and as "What does this teach us?". It's honestly almost a perverse question to primarily ask for something that aims to have practical value.
The clear reason this has worth is that it's much faster than TSC.
HN is going to be a weird place to be between people loving AI and the loss of loving tech. I don’t remember last time I watched a good tech dev video. At least the video game industry is a little bit shielded from this but soon it will be the same when we see the speed AI is advancing.
I don't see at all that this is stifling love of tech. Sure if you liked to engage in language wars or vim vs emacs or tabs vs spaces then yes that's over. But tech as we know it is currently exploding, a ton of stuff that simply was not practically possible before is now possible.
The point is that it's clearly not just a less maintained, less tested implementation. It's 1.61x faster for one in real world apps, in some case even almost 4x faster.
I am curious why it's faster. Golang isn't that much slower than Rust, aside from the garbage collector. 4x performance improvements is much bigger than what I expected.
This was a lot of tokens to spend on a port. I wonder if all of the porting activity out there would be better supported by investing in really good deterministic automatic translators that do 90% of the job cheaply and delegate any decisions that have to be made back out to the AI agent?
I have to wonder if LLM reimplementations verified against years of human tests will have holes where original authors thought that tests are not necessary and common sense is enough.
I have to wonder this to preserve my ego as a human. I wonder have to this because I sure as hell am never going to go through all that code to check.
Of course they will. For about 3 days, and then it will get fixed.
At this point, for software which has an extensive test suite (or a large body of interoperable software), humans are more likely to create these kinds of gaps than the LLMs are.
LLMs have been trained to fix compile errors in a loop. Give them years of human labor worth of tests and they will fix compile errors until it works (as well as the tests can ensure). Now you have a codebase no one has read. Good luck adding new code to it.
Let's also not forget that Bun was claiming a rewrite in 11 days or whatever it was, but actually spent three months of human labor fixing hundreds of issues their rewrite introduced before shipping it as a release.
I'm curious how much unsafe this uses? Are these ports making idiomatic code or just using unsafe all over. There doesn't seem to be anything in the readme.
As much as possible mechanical (non-LLM) verification of equivalence that generates good diagnostic messages, mechanical translation steps that generate good diagnostic messages, a good memory system to reduce rework... (edit: and mechanical coverage measurement)
Probably zero, or even a profit. The idea that these companies are making a loss on subscriptions is an unsubstantiated HN fallacy.
On the contrary, they're probably making a small profit on subscriptions, or at least break even, and absolute bank on API pricing. Anthropic recently reported an 80% gross profit margin, for example.
It might, but also might not. I would imagine the economics are genuinely very tangled up even before you start obfuscating it - it’s very hard to fairly allocate fixed costs between training and inference.
I think a clearer picture would be from inference-only orgs.
Sibling dead comment said it already but that's likely an estimate of what it would cost via API and in practice it was a few hundred or thousand dollars of subscription fees.
It is beginning to look like the future is treating Rust as an optimization pass. Write your program in a higher-level language like Typescript, Go, Python, etc. and then have LLMs compile that to Rust.
Not really, the future are natural languages, with some markdown files for formal specification, yaml or whatever, and then Assembly gets generated directly.
3 GLs are going to slowly fade out as LLM tooling improves, and coding looks increasingly as envisioned by 4 GLs advocates.
> Spent over $400,000 in API priced tokens with GPT-5.6 Sol and GPT 6 Astra
Tangent here, but I think this bit is super interesting!
You could viably hire someone to do this work for that kind of money - I think the interesting thing is that substantially less interested/experimenting engineers would consider paying for a human to do this work, than would happily chuck a big amount of money into an LLM.
I don't have any suggestion about why that exists, but it's a strange and interesting contract.
This is Theo's project, and he has been very transparent about how he is only doing these things because api prices are heavily subsidized by subscription pricing. The reason people aren't willing to pay 400k to an engineer to do this work is because they aren't paying the AI this much.
I don't think I've seen an enterprise paying API prices doing these sorts of rewrites, and honestly, I think until that happens I will remain skeptical about LLM AI's viability as a profitable business.
I found the whole section super interesting... I'll copy it here:
I used a lot of OpenAI models to try and complete this port. In total I did over $400,000 in API priced tokens with GPT-5.6 Sol and GPT 6 Astra. They wrote over 1.3m lines of Rust over multiple months of /goal loops and never got past like 84% compat.
When I saw how little my Claude Code limits were burning, I figured it'd be fun to throw Opus 5.5 at this. It had a working v0 in 10 hours.
I assumed it kept using the code the Codex models wrote. I was wrong. Opus 5.5 started from scratch. It got further than Astra in 1/10th the time.
I let it keep going, and it definitely did. Total token spend was ~$24,047 of API spend over 2 weeks. I was using my Claude accounts, and it worked out to somewhere between 925% and 983% of my $200 plan weekly limits.
Expensive, for sure, but not that bad considering how much work has went into typescript-go.
Great advertisement for Claude Code...
I've had the same thought about the maths - they've spent millions on tokens, I can't recall anyone spending millions on mathematicians.
Paraphrasing from README.md:
>Spent over $400,000 in API priced tokens with GPT-5.6 Sol and GPT 6 Astra. They wrote over 1.3m lines of Rust over multiple months of /goal loops and never got past like 84% compat.
Then they let Claude code lose on the problem:
> I figured it'd be fun to throw Opus 5.5 at this. It had a working v0 in 10 hours. I assumed it kept using the code the Codex models wrote. I was wrong. Opus 5.5 started from scratch. It got further than Astra in 1/10th the time.
In the warnings from README.md
> Also worth mentioning: I've never read a line of this code.
They spent over $400k for an experimental project and published it without reading a single line of code??
I am so confused about the motivation behind this.
TypeScript team member here - this is impressive work, and it's honestly crazy that 3 of these ports have popped up in the last week! I'm currently AFK but we're hoping to learn more and we'll have more to say on this soon.
Kudos for keeping up! Golang seemed like a very reasonable choice when you ported there originally but the world is changing _fast_.
Why do you think Golang is no longer a reasonable choice?
I suppose my comment sort of implies this. I think the naive knee jerk reaction is that if you can spit out perfect machine level code against a specification, then logically being closer to bare metal would help with performance. However, without knowing the details of how things worked it is tough to say whether it is truly "reasonable" or not.
But more importantly, the Typescript team themselves [1] picked Golang mostly out of ergonomics and ease of porting at the time. It is becoming more increasingly more to "taste" what ergonomics means. Notably, they did not pick it because they believed it to be the fastest option. But speed was on the mind, giving way to ergonomics and ease of porting. At least, this is my read.
[1] https://github.com/microsoft/typescript-go/discussions/411
The author of esbuild used Rust and later switched to Go. He describes some of his experience here:
https://news.ycombinator.com/item?id=22336284
It is one person and one project, but I found it interesting to read about his experience.
I like Go, but I probably would have tried Rust first because I want pattern matching when implementing languages.
Presumably because Rust is more suitable than Go in agentic coding? (I'm not sure, but that seems the purported answer.)
Language and compiler design/ implementation is very much a human task. There's lots of nuances to consider. I doubt they want agent spaghetti code in their codebase either.
Rust is more close to metal. No GC
GC doesn’t affect performance (at least visibly), memory allocations on critical paths do. Moreover, Rust ARC may cause memory fragmentation and perform worse than a GC.
Writing Rust, Zig, or Go, you would still control memory manually, where it matters.
It does have an impact. Discord blogged about unavoidable GC delays in go that occurred because go needed to traverse live memory, their only option (in go) for reducing that pause was to reduce the size of their data set.
There is no connection. Rust has ARC which performs worse than the tracing GC.
Than Go's GC? I think that is fairly unlikely in most cases.
It depends on the workload. ARC can be more expensive than tracing GC, especially with heavily shared objects.
But my main point was that the presence of a GC has nothing to do with how close a language is to the metal. Memory management strategy and low-level capabilities are two separate things.
Rust is better at targeting WebAssembly
WASM makes no sense and has no future.
It does here, https://www.typescriptlang.org
I’m not saying it doesn't work. I’m saying it’s a waste of resources and won’t gain widespread popularity. The current trend is porting source code to languages that generate native code.
I dislike the WebAssembly advocacy of some folks, where they act as if there weren't similar attempts all the way back to UNCOL in 1958, and even in recent history, seem to only focus on JVM, while ignoring the CLR and BEAM.
However, there are indeed some cases where it does make sense since we have it around anyway, and one of them is running the Typescript compiler in the browser, now that it is written in a compiled language.
Languages that target WebAssembly _are_ languages that generate native code? C & Rust for example.
why would that be important? Having a typescript compiler in wasm seems like a pretty niche need.
Because you need to keep the compiler running here,
- https://www.typescriptlang.org
- https://code.visualstudio.com/docs/remote/vscode-web
Theo is making a new twist on the idea of IDE which hosts the application itself. I'm sure there's going to be a video about it soon. But the idea is to embed the whole compilation and type-check in wasm in the browser which then runs the app immediately, as the agents edit the code.
Why?
Wasm is a first-class target for Rust, which produces faster and smaller Wasm binaries. My understanding is that you need to use TinyGo to reasonably target Wasm with Go.
Go having a garbage collector isn't necessarily bad, since Wasm has GC, but I think Go's GC can't easily use Wasm GC because Go has interior pointers.
WASM is also a first-class target for Go (gc).
However, it has a relatively large runtime, so TinyGo is often preferred when bundle size needs to be small.
I think the main advantage is size. Rust doesn't need to include the runtime.
I'm assuming they mean because it doesn't need a garbage collector or much of a runtime at all. You can target direct to WASM and use I guess WASI, while Go brings with it a garbage collector and so on.
Was it ever a reasonable choice? At the time the choice was made the tooling ecosystem was already split between Rust and JS, and by putting TS in Go they chose to split it three ways. Why not split it 4 or 5 ways then? The pressure will always be towards less duplicated work.
The only real choice of language to build the next generation of JS tools in is JS. Anything else is a vote of no confidence in ourselves.
It's ok for a language (JS in this instance) to be good in a domain but not good in all domains. Pulling a language in every direction forces it to make compromises that hurt its applicability in specific areas.
Now I want to see somebody to port it to c# :)
I think the future of this is -- get your AI agent to investigate this port for anything actually worthwhile, comparing it to the go version, and then take those good ideas and actually maybe bring them into the real compiler.
Maybe. Who knows, it's all changing so fast!
What a waste of power... $424.000 of API priced tokens amounts to how many kWh burned? And how would you trust this to build your code? Motivations says it all:
Motivations
Maybe there are just so many of these but at this point for vibe coded stuff like this I’m just like “who cares.” LLMs can make stuff like this now. Unless this reaches some critical mass of usage among developers why would I use it? It’s just a less maintained, less tested implementation. The more important question is “what does this teach us?” And idk the answer to that one.
For me it removed all the fun to play with whatever projects are announced in HN, now it is all AI generated, why bother.
It teaches us the Rust community continues to be relentless in their efforts to rewrite virtually everything in Rust.
Isnt the guy who made this a typescript developer?
Yes. This has nothing to do with people trying to rewrite things in Rust for fun. It's a part of his development project that I'm sure will get a video in the future.
I think the difference here is that the original motivation for choosing Go over Rust was that Go was easier to port TypeScript to.
LLMs change the calculus a lot, the result seems faster, and it's possible that the TypeScript team might want to change directions. It'd be a big deal, but I bet they'll at least discuss it.
There is a long interview with Anders Hejlsberg here https://www.youtube.com/watch?v=cywK3XYYJ2o&pp=ygUbYW5kZXJzI... where he talks about that, in short from what he said 1-2 months ago, I doubt that they will change directions.
Generally building something is not about "What does this teach us?" but rather "What can this be used for?". You don't build a house and as "What does this teach us?". It's honestly almost a perverse question to primarily ask for something that aims to have practical value.
The clear reason this has worth is that it's much faster than TSC.
HN is going to be a weird place to be between people loving AI and the loss of loving tech. I don’t remember last time I watched a good tech dev video. At least the video game industry is a little bit shielded from this but soon it will be the same when we see the speed AI is advancing.
No more joy crafting things as a dev…!
I don't see at all that this is stifling love of tech. Sure if you liked to engage in language wars or vim vs emacs or tabs vs spaces then yes that's over. But tech as we know it is currently exploding, a ton of stuff that simply was not practically possible before is now possible.
I guess I’m just perverted then.
That's why the commenter said
> It’s just a less maintained, less tested implementation
Then they asked if it teaches something because it is a risky utility to them. You're trying to make it sound a lot weirder.
The point is that it's clearly not just a less maintained, less tested implementation. It's 1.61x faster for one in real world apps, in some case even almost 4x faster.
I am curious why it's faster. Golang isn't that much slower than Rust, aside from the garbage collector. 4x performance improvements is much bigger than what I expected.
If that was your point, why did you spend a whole paragraph saying "teach us? What a perverse question."
The comment I responded on put forth a generic argument that implied if we can not learn from it it has no value. That is what I was responding to.
I wish Anthropic/OpenAI would give me free tokens so I could make random things like this
I already pay for subs on both but the tokens are already spoken for with other projects
This was a lot of tokens to spend on a port. I wonder if all of the porting activity out there would be better supported by investing in really good deterministic automatic translators that do 90% of the job cheaply and delegate any decisions that have to be made back out to the AI agent?
I have to wonder if LLM reimplementations verified against years of human tests will have holes where original authors thought that tests are not necessary and common sense is enough.
I have to wonder this to preserve my ego as a human. I wonder have to this because I sure as hell am never going to go through all that code to check.
Of course they will. For about 3 days, and then it will get fixed.
At this point, for software which has an extensive test suite (or a large body of interoperable software), humans are more likely to create these kinds of gaps than the LLMs are.
LLMs have been trained to fix compile errors in a loop. Give them years of human labor worth of tests and they will fix compile errors until it works (as well as the tests can ensure). Now you have a codebase no one has read. Good luck adding new code to it.
Let's also not forget that Bun was claiming a rewrite in 11 days or whatever it was, but actually spent three months of human labor fixing hundreds of issues their rewrite introduced before shipping it as a release.
In the real world project section, compile times vs tsc7 were 1-3s less. Sometimes the percentages are far more impressive than absolutes.
"In total I did over $400,000 in API priced tokens with GPT-5.6 Sol and GPT 6 Astra."
wut?
This post should be titled “Rust port of Go rewrite of Typescript compiler”
I'm curious how much unsafe this uses? Are these ports making idiomatic code or just using unsafe all over. There doesn't seem to be anything in the readme.
Have a look yourself: https://github.com/search?q=repo%3Apingdotgg%2Fts-rust+unsaf...
Doesn't look like any on skim, also looks like an explicit goal was to avoid unsafe code: https://github.com/pingdotgg/ts-rust/blob/ad8f2746ea85a6354b...
Are there any specific harnesses or techniques to aid in porting stuff to Rust? Specialized in terms of token efficiency
As much as possible mechanical (non-LLM) verification of equivalence that generates good diagnostic messages, mechanical translation steps that generate good diagnostic messages, a good memory system to reduce rework... (edit: and mechanical coverage measurement)
Wait - you spent north of $400,000?
People love to flex by using the API costs for their subscription tokens.
I personally find it ridiculous. The cost is what you paid, not what someone else could have paid. Markets and all that.
It would be like using spot instances on AWS but flexing by boasting about how much on-demand $$ compute you used.
In this case, I think they are just giving you an idea of what it cost to do this. They aren't saying it's prescriptive
Except.. It doesn't really give you an idea what it cost? That's not what it cost. How much did it actually cost? Who knows!
It's also TheoGG sooooo.. Those influencer instincts tho.
Assuming he used $400,000 in tokens, but used subs
I wonder how much AI companies lost on this?
Edit: not that I’m worried about them, I’m just curious
> I wonder how much AI companies lost on this?
Probably zero, or even a profit. The idea that these companies are making a loss on subscriptions is an unsubstantiated HN fallacy.
On the contrary, they're probably making a small profit on subscriptions, or at least break even, and absolute bank on API pricing. Anthropic recently reported an 80% gross profit margin, for example.
https://www.reuters.com/business/retail-consumer/anthropic-t...
Profitable, if you just ignore how much money they spend
Excited for their IPO just to put this to bed, one way or another.
It might, but also might not. I would imagine the economics are genuinely very tangled up even before you start obfuscating it - it’s very hard to fairly allocate fixed costs between training and inference.
I think a clearer picture would be from inference-only orgs.
What are “subs”?
Subscription
So you pay $20/mo, let it run for a long time, then claim you used the equivalent of $1/2 million?
Most of it was wasted on OpenAI tokens. It actually cost them $24k USD according to the github description.
So, you could do it for a couple hundred bucks using DeepSeek.
No that's the theoretical token value cost based on chaining multiple 20x subs.
An absurd amount. The cost wouldn't exceed $500 under a subscription model.
Sibling dead comment said it already but that's likely an estimate of what it would cost via API and in practice it was a few hundred or thousand dollars of subscription fees.
$420,000 worth of tokens, though I'm pretty sure he used subs.
> I did over $400,000 in API priced tokens
i'd bet this is a few subscriptions over a few months rather than paying directly for tokens.
I wonder how that cost compares to paying a human programmer to produce the same end result.
imagine if they had the willingness to re-write it for human consumption. might a well delete it
Is this going to be a serious project? No.
Is this going to be maintained? Nope
Nevertheless this and many others are showing what is possible. Much akin to how someone ports doom onto a microwave.
We'll look back at this time with great fondness when we collectively discovered a whole new gear.
> Nevertheless this and many others are showing what is possible.
More unserious and unmaintained projects at 400k a pop. Sick.
I honestly have no idea what the future of tech will look like at this point. Too much has changed.
It is beginning to look like the future is treating Rust as an optimization pass. Write your program in a higher-level language like Typescript, Go, Python, etc. and then have LLMs compile that to Rust.
Not really, the future are natural languages, with some markdown files for formal specification, yaml or whatever, and then Assembly gets generated directly.
3 GLs are going to slowly fade out as LLM tooling improves, and coding looks increasingly as envisioned by 4 GLs advocates.
Why does it seem like this is falling off the front page really fast? There's older posts with less points that are way higher.