Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

We tasked Opus 4.6 using agent teams to build a C Compiler

anthropic.com

Post reply on HN
Print view View original

251–260 of 787 posts

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#251

Post by seba_dos1 »

RyanCavanaugh wrote: post #103
halxc wrote: post #99

Earlier quoted context omitted.

We all saw verbatim copies in the early LLMs. They "fixed" it by implementing filters that trigger rewrites on blatant copyright infringement. It is a research topic for heaven's sake: https://arxiv.org/abs/2504.16046

The internet is hundreds of billions of terabytes; a frontier model is maybe half a terabyte. While they are certainly capable of doing some verbatim recitations, this isn't just a matter of teasing out the compressed C compiler written in Rust that's already on the internet (where?) and stored inside the model.

> The internet is hundreds of billions of terabytes; a frontier model is maybe half a terabyte.

The lesson here is that the Internet compresses pretty well.

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#252

Post by rishabhaiover »

hmry wrote: post #248
rishabhaiover wrote: post #233

Earlier quoted context omitted.

[flagged]

Just tired of AI companies having more rights than natural people when it comes to copyright infringement. Let us have some of the fun too!

I apologize for making that assumption.

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#253

Post by bopbopbop7 »

ndesaulniers wrote: post #243
bopbopbop7 wrote: post #228

Earlier quoted context omitted.

A pay to use non-deterministic compiler. Sounds amazing, you should start.

Some people care more about compile times than the performance of generated code. Perhaps even the correctness of generated code. Perhaps more so than determinism of the generated code. Different people in different contexts can have different priorities. Trying to make everyone happy can sometimes lead to making no one happy. Thus dichotomies like `-O2` vs `-Os`. EDIT (since HN is preventing me from responding): > S…

Some people care more about compiler speed than the correctness? I would love to meet these imaginary people that are fine with a compiler that is straight up broken. Emitting working code is the baseline, not some preference slider.

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#254

Post by shakna »

ndesaulniers wrote: post #216
shakna wrote: post #200

Earlier quoted context omitted.

> Opus was unable to implement a 16-bit x86 code generator needed to boot into 16-bit real mode. While the compiler can output correct 16-bit x86 via the 66/67 opcode prefixes, the resulting compiled output is over 60kb, far exceeding the 32k code limit enforced by Linux. Instead, Claude simply cheats here and calls out to GCC for this phase Does it really boot...?

> Does it really boot...? They don't need 16b x86 support for the RISCV or ARM ports, so yes, but depends on what 'it' we're talking about here. Also, FWIW, GCC doesn't directly assemble to machine code either; it shells out to GAS (GNU Assembler). This blog post calls it "GCC assembler and linker" but to be more precise the author should edit this to "GNU binutils assembler and linker." Even then GNU binutils contai…

Yeah, didn't mention gas or ld, for similar reasons. I agree that a compiler doesn't necessarily "need" those.

I don't agree that all the claims are backed up by their own comments, which means that there's probably other places where it falls down.

Its... Misrepresentation.

Like Chicken is a Scheme compiler. But they're very up front that it depends on a C compiler.

Here, they wrote a C compiler that is at least sometimes reliant on having a different C compiler around. So is the project at 50%? 75%?

Even if its 99%, thats not the same story as they tried to write. And if they wrote that tale instead, it would be more impressive, rather than "There's some holes. How many?"

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#255

Post by HDThoreaun »

lossolo wrote: post #111
Philpax wrote: post #61

Earlier quoted context omitted.

What Rust-based compiler is it plagiarising from?

Language doesn't really matter, it's not how things are mapped in the latent space. It only needs to know how to do it in one language.

Ok you can say this about literally any compiler though. The authors of every compiler have intimate knowledge of other compilers, how is this different?

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#256

Post by chasd00 »

ndesaulniers wrote: post #243
bopbopbop7 wrote: post #228

Earlier quoted context omitted.

A pay to use non-deterministic compiler. Sounds amazing, you should start.

Some people care more about compile times than the performance of generated code. Perhaps even the correctness of generated code. Perhaps more so than determinism of the generated code. Different people in different contexts can have different priorities. Trying to make everyone happy can sometimes lead to making no one happy. Thus dichotomies like `-O2` vs `-Os`. EDIT (since HN is preventing me from responding): > S…

a compiler introducing bugs into code it compiles is a nightmare thankfully few have faced. The only thing worse would be a CPU bug like the legendary Pentium bug. Imagine you compile something like Postgres only to have it crash in some unpredictable way. How long do you stare at Postgres source before suspecting the compiler? What if this compiler was used to compile code in software running all over cloud stacks? Bugs in compilers are very bad news, they have to be correct.

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#257

Post by Aurornis »

bopbopbop7 wrote: post #228
beambot wrote: post #221

Earlier quoted context omitted.

This is getting close to a Ken Thompson "Trusting Trust" era -- AI could soon embed itself into the compilers themselves.

A pay to use non-deterministic compiler. Sounds amazing, you should start.

Application-specific AI models can be much smaller and faster than the general purpose, do-everything LLM models. This allows them to run locally.

They can also be made to be deterministic. Some extra care is required to avoid computation paths that lead to numerical differences on different machines, but this can be accomplished reliably with small models that use integer math and use kernels that follow a specific order of operations. You get a lot more freedom to do these things on the small, application-specific models than you do when you're trying to run a big LLM across different GPU implementations in floating point.

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#258

Post by defen »

falloutx wrote: post #96
Philpax wrote: post #85

Earlier quoted context omitted.

No human developer, not even Fabrice Bellard, could reproduce this specific result in a week. A subset of it, sure, but not everything this does.

just forked https://github.com/Vexu/arocc and it took me 5 seconds to complete it.

That can't build the Linux kernel though.

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#259

Post by Calavar »

raincole wrote: post #161
geraneum wrote: post #36

Earlier quoted context omitted.

> This was a clean-room implementation This is really pushing it, considering it’s trained on… internet, with all available c compilers. The work is already impressive enough, no need for such misleading statements.

It's not a clean-room implementation, but not because it's trained on the internet. It's not a clean-room implementation because of this: > The fix was to use GCC as an online known-good compiler oracle to compare against

The classical definition of a clean room implementation is something that's made by looking at the output of a prior implementation but not at the source.

I agree that having a reference compiler available is a huge caveat though. Even if we completely put training data leakage aside, they're developing against a programmatic checker for a spec that's already had millions of man hours put into it. This is an optimal scenario for agentic coding, but the vast majority of problems that people will want to tackle with agentic coding are not going to look like that.

Re: We tasked Opus 4.6 using agent teams to build a C Compiler

#260

Post by Aurornis »

halxc wrote: post #99
antirez wrote: post #88

Earlier quoted context omitted.

The LLM does not contain a verbatim copy of whatever it saw during the pre-training stage, it may remember certain over-represented parts, otherwise it has a knowledge about a lot of things but such knowledge, while about a huge amount of topics, is similar to the way you could remember things you know very well. And, indeed, if you give it access to internet or the source code of GCC and other compilers, it will imp…

We all saw verbatim copies in the early LLMs. They "fixed" it by implementing filters that trigger rewrites on blatant copyright infringement. It is a research topic for heaven's sake: https://arxiv.org/abs/2504.16046

Simple logic will demonstrate that you can't fit every document in the training set into the parameters of an LLM.

Citing a random arXiv paper from 2025 doesn't mean "they" used this technique. It was someone's paper that they uploaded to arXiv, which anyone can do.

Post reply on HN