srcreigh

Astra was always lazy as a default. I wrote about this >1 month ago https://srcreigh.ca/posts/gpt-6-astra-wont-try-hard-unless-y...

What did you try to make generalized harness, and what kind of failures did you encounter?

> Generally

Except for Postgres

Not if ie6 is your only option

Agree that it should be easier to test on safari from other platforms.

But no way is it a good idea to let other browser engines onto the OS.

I can’t make it website owners problem that their site doesn’t work on non-Google browser. Apple can.

Safari sucks but it’s the only thing stopping Chrome from taking over everything.

Im happy if Google never has that amount of control. We don’t know how bad it could have been, we have enough privacy issues already.

They might be used to Postgres, which doesn't have primary index.

> Part of me would like a little longer in the world where the problem is still open and I am still looking for its solution. But reaching a summit, even by someone else’s route, comes with a view. From here I can see new mountains, and I look forward to climbing them with my students, collaborators and the machines.

https://dakshitakhurana.substack.com/p/classical-at-heart

Cool! Also .aa and .zz based on the table from your Wikipedia link.

It’s not so hard to get root shell on some routers via the admin panel, which usually has the same password as the wifi network or a default password. From there the device can capture dns logs.

This is incredible, thanks a lot for making it available for us.

The "new data" is: person A prompts an LLM to create or modify a tool, person A distributes code person B, person B's LLM uses the tool via docs/help/error. That is a direct path for an LLM to "know more" from a human than what's in its training data.

If we get a new programming language not in the training dataset, we could give an LLM a decent compiler with compile errors, and some sample code and it would be able to write code in the new language without training.

setting aside the dynamics question, it is known that LLMs have limits wrt the theorems they can prove without human input, see my comment here: https://news.ycombinator.com/item?id=49756412

LLMs receive new data via input context, not just training data.

Thought experiment: How effective will 2026 LLMs be for humans in 2526?

It's not game over just because 500 years are missing from the training data. The important question is how well can 2526 humans make culture and knowledge navigable to LLMs via tool calls.

Today's LLMs might need for example sub agents to translate to 2526 English, sub agents to read 2526 docs.

It's _really not clear_ whether 2026 LLMs will be useless. To believe that reflects an enormous misunderstanding.

I agree. I believe it could be the best piece about AI ever written. The themes of lost dreams reawakened, sobering up of egos, sci fi technological development, the role of humanity. It’s a weird feeling to see that it’s actually really happening.

I like to use firecracmer, since it has no support for many things, eg host filesystem mounting in the guest.

Can you make any similar guarantees about Drop?

Which models did you try for which tasks?

> You'd have to test batching of multiple thinking streams in order to try and raise overall tok/s via layer-wise reuse of the streamed weights

isn’t this very straightforward to do..? I thought batching for Qwen models is already proven out.

> but this would decrease single-session performance even further

Well let’s take Qwen 3.8 27B. Throughput for M3 at 8 agents is 4x compared to single agent. [1]

It’s really not clear to me that 8 concurrent agents at half speed will be worse task completion latency than 1 agent.

And that’s M3 studio benchmarks, not even M5 ultra, and without the many software improvements we will see

If you haven’t tried Qwen 3.8 27B xhigh on a task you might not get the hype. Idk.

If you’ve tried doing this and don’t like it sure, and be specific about what isn’t effective, but let’s not speculate.

[1]: https://omlx.ai/benchmarks/performance/69kzkrv8?utm_source=c...

I think the issue is generalization, if you were more specific about which local models aren’t good enough for which tasks compared to which frontier models in your experience, it’d be a lot more informative

If your networking stack and filesystem and who knows what else are exposed to the guest, its not isolation.

It’s better than Docker, but it can’t be compared to Firecracker at all. Firecracker actually minimizes the attack surface whereas smolvm does not

This is great as a first look, but the author is not a developer, so we don't yet know whether a dev can be as productive with local models on M5 Mac Studio compared to a 20x subscription plan.

I'm also curious about any new low hanging optimization opportunities in the kernels for this new hardware.

It's already clear to me that M5 Mac Studio is more cost-effective than anything you can run on open router, assuming decent utilization.

The M5 Mac Studio will be the most cost effective way to run uncensored cyber capable open agents.

An exciting tipping point will be if programmers can get an Astra-Ultra like experience all week with this hardware. That would be a real sense where this hardware exceeds the value of even 20x cloud subscriptions.

Supports direct filesystem access and who knows what else. It is unsuitable for running untrusted code on Linux.

This post explains how a GPT agent broke out of qemu VM. It could not break out of firecracker.

https://blog.trailofbits.com/2026/08/26/vms-wont-contain-cyb...

Why? Firecracker mounts very few host systems into the VM, exposing minimal host code to malicious guests. Qemu and smolvm expose much more.

So yeah, smolvm is more like a docker or qemu alternative, definitely useful but NOT relevant to the discussion of sandboxing malicious code

I can launch Astra to deploy changes to my homelab via creating Forgejo issues from my phone. That kind of system is pretty hard to set up without kubernetes.

Even if you confine yourself to a dev workstation, having 5 agents concurrently building testing deploying code makes your computer loud and/or hot.

K8s jobs don’t run in a secure runtime. K8s jobs don’t give you dynamic zero trust permissions scopes. Restoring a harness in 500ms is really fast, much faster than naively creating a new job downloading session and ‘codex resume’ etc.

Tasks are a good scope for zero trust permissions

I have 6 and ended up needing to use my gaming PC for a build server.

I think you could get by with 1 computer, but it’ll have to have a pretty decent machine.

Between agents running tests, CI, docker image builds, an average $400 mini PC won’t cut it.

Don’t forget also many older mini PCs don’t support KVM. Some newer ones don’t support AVX/ mongodb.

It’s not so easy to buy any old hardware sadly.

It doesn’t provide isolation. It is not even part of the conversation.