MallocVoidstar

Yeah, I used to get stuff directly from China, then after the postage rate changes I was getting packages that had gone China -> Thailand -> Azerbaijan -> USA. Nowadays they seem to batch packages for the good shipping, from Aliexpress in China to (I assume) some US subsidiary, and then from there it gets parceled out to a shipping company seemingly at random (Maersk, Amazon, USPS).

Claude Code 2.1.207 and OpenCode 1.17.18, both pinned to claude-sonnet-4-5

So not only is this article AI-written, but the testing was entirely done by AI, too? I can't see any other reason to use such an old model.

Our traffic passes through a local LLM gateway that wraps requests in its own envelope, a constant we measured at roughly 6,200 tokens with bare calibration requests

Why do you need to do calibration requests to figure out how your own gateway is affecting requests?

Its subagent lane did not complete cleanly through our gateway

We attempted to toggle extended thinking in both harnesses and are declining to publish numbers. Our gateway applies its own thinking policy, neither harness's toggle demonstrably survived the path, and anything we quoted would be noise.

Why is your own gateway screwing with your testing?

For one thing, CF is probably the largest entity serving pirated content internationally while hiding the identities of actual perpetrators for privacy.

That's awesome, glad to hear it

Some of them will publicly break it for whatever reason, an NVIDIA engineer (@blelbach) recently made some posts on X about his results using 5.6 Sol Ultra to optimize stuff and then deleted them shortly after (presumably got yelled at).

There's something like 20 companies with access.

No, depending on the complexity of the issue models can be into loops, where they go "this is definitely an issue and must be fixed", and then the resulting fixed code gets "this is definitely an issue and must be fixed", and then the resulting fixed code has the original 'issue'.

I don't want any random program on my computer to be able to change all of my system settings and terminate security processes.

The preceding sentence is

Our safety assessments found that Sonnet 5 shows an overall lower rate of undesirable behaviors than Sonnet 4.6, and is generally safer to use in agentic contexts.

which is obviously painting that as a good thing. So reading the next sentence as "in other good news" is reasonable.

There are a lot of companies reselling Claude to Chinese users. You use their base URL but it's still going to Anthropic.

OpenAI is saying they want it to be temporary, not that it will be.

"Mobile" generators that are sitting in one place and not moving.

Memphis is majority black

Do you think everyone on HN sees every submission?

The vast majority of the complaints I see about "[LLM] is worse now" never include any examples.

I'm not reading eight thousand AI-generated words saying one single person is ruining every model.

https://nitter.net/PeteHegseth/status/2065897156226015690

You think that if someone can get a model to write a beginner's guide to exploiting code that requires writing your own purposefully vulnerable program, then the creators of that model should be arrested?

Anthropic refused to allow the US Government to conduct mass surveillance, which made the US Government mad. OpenAI was fine with it as long as it was 'legal' mass surveillance. OpenAI is not going to get banned, even if their next model is both better and more dangerous than Mythos.

It's only rewarding hype if the ban gets dropped. If "foreign Anthropic employees that live in the US can't use Fable/Mythos" stays it harms them, if they don't drop the ban and Fable/Mythos stay limited to "every single person who uses the model must individually provide their ID to prove American-ness" it harms them.

Google shows completely wrong timestamps all the time. I'm pretty sure they just randomly grab vaguely date-like text from pages and declare it the date the page was created.

Some are made in China and Japan, most are made in Wales.

https://www.jeffgeerling.com/blog/2023/how-raspberry-pis-are...

It's useful if you need GPIO but not $350 useful. Nowadays you can get used office mini PCs with a 10th gen Intel and 16GB RAM for like $200 and they'll come with an SSD. No idea why anyone would buy an expensive Pi.

At least get a -C6 if you're trying to replace a Pi with a microcontroller.

Opus 4.0 and 4.1 are more expensive than Fable.

AI (assisted?) summary of the March 2025 Anthropic paper https://transformer-circuits.pub/2025/attribution-graphs/bio...

The article is from May 29 2026, they're lying about that update being 'recent' and coming after the article to make themselves look better.

"As explained by" someone regurgitating the views of the "Citizens’ Council for Health Freedom", an anti-vax, right-wing organization that opposes HIPAA, Medicare, and government involvement in healthcare in general.

It is entirely possible that any released model access will be heavily guardrailed against hacking attempts

Yes, Anthropic have said they made Opus 4.7 worse at this on purpose.

It is entirely possible that Mythos is a different architecture or size

It has 5x the token pricing of Opus 4.7, so it's probably larger.

https://pbs.twimg.com/media/HItbXhvW4AAMD8W?format=jpg&name=...

All of their repos have been copied and are up for sale. Attackers are TeamPCP, the creators of the Shai-Hulud malware.

In addition to people pointing out your LLM got the pricing wrong,

The pricing and "including thinking tokens" framing position it as a reasoning-capable flash model rather than just a pure speed optimization

Every Gemini model starting with 2.5 has been a reasoning model.