I think they just want to be a winner in the “next thing.” They hit social networking, but missed mobile operating systems and didn’t compellingly win at social media. Eventually an ambitious person with a bazillion dollars wants a clear win, right?

We shouldn’t try to build a worse version of a human. We should try to build a better compiler and encyclopedia.

We tried that. It was called Cyc. It never got even close to the level of capabilities a modern LLM has in an agentic harness — even on common sense and reasoning problems!

That sounds like a "get wealthy slowly" plan, while the LLM prophets are more focused on "get rich quick".

The compare against “DeepSpeed ZeRO-3” apparently.

FWIW Zero-3 refers to a common strategy for sharding model components across GPUs (commonly called FSDP-2, Full Sharded Data Parallel). The "3" is the level of sharding (how much stuff to distribute across GPUs, e.g. just weights, versus optimizer state as well, etc.)

IMO there’s room for something more recent, maybe a Titan or something, to stand in as an avatar for making GPUs as compute accelerators a thing. I know that’s been going on forever, but at some point it went from some niche hacky thing to a primary use-case for the cards.

But yeah this list has a on of incremental bumps on it. Maybe there was some mixing of cards that mattered historically and cards that mattered to the author.

Are agents something special? We already had LLMs that could call tools. Agents are just that, in a loop, right?

Yes, and this is an incredibly powerful idea. Running fluid flow simulations inside an optimization loop (monte carlo + gradient descent) revolutionized aircraft design, nuclear simulations and geophysics. When the tool being called updates the LLM's training data, or runs experiments that the LLM can learn from, then that potentially becomes a self-improvement loop.

Roughly speaking - yes. Still, it's an advancement - even if it's a small one - on the usual chatbots, right?

P.S. I am well aware of all of the risks that agents brought. I'm speaking in terms of pure "maximum performance", so to speak.

Does Firefox really not unload the tabs in that case?

It does. You can also do it by hand via the right-click on tab menu

It is not actually an issue. The article isn’t based on any technical aspects of the OS, just the reported system requirements.

Maybe the bots should be made to write MISRA-C. It isn’t like they get annoyed, right?

I wonder, is there a way to only request reformulations that don’t involve branches? The tool already seems quite nice, but that might be a good feature.

Also, I’m not sure I understand the speedup. Is it latency or throughput?

Author here. The speed up is modeled throughput, though the model is relatively naive. It's possible to disable branches by turning off the regimes flag, see https://herbie.uwplse.org/doc/1.0/options.html

Nice!

What’s uwplse mean? I mixed up the letters and misread it as ulp-wise which works for the project, haha.

University of Washington Programming Languages and Software Engineering (research group).

I'm not at UW any more, I'm now at Utah, but some of the Herbie team is at UW and they provide the infrastructure

LibreOffice is a pretty bad name, it is too clearly a spin-off of OpenOffice and never really gained its own identity. Being identifiable as a bad project’s better fork is kind of a weak starting position.