undefined

points

by redox995 hours ago |

comments

by Rapzid4 hours ago|

[-]

Yeah, I use Astra for destroying vaguely scoped asks and tasks, and then for high-level design and plan generations..

Otherwise I'm using 5.6 Sol for actual plan execution and review..

by amluto45 minutes ago|

prev|

[-]

I use Astra for rapidly consuming my token limit on a task that would not consume it on 5.6 Sol.

(I have not done anything quantitative here. For one thing, OpenAI’s billing pages and the codex-rs frontend make it pathetically difficult to get any real data. Some day I should wire up a proxy to extract actual stats.)

by cmrdporcupine5 hours ago|

prev|

[-]

Yeah.

Astra was/is superior for planning type tasks. It was capable of doing seemingly magic things with rather vague/lazy instructions ("I need to be able to test this on Windows, maybe a qemu VM or something? Shrug." ... 1 hour later "yeah i built you a whole qemu + eval windows image + harness of powershell scripts + shell scripts to retrieve & verify harness.").

And for UI work -- which is not something I do a lot of but do here and there -- it was clearly superior to 5.6 Sol.

But it also feels sloppier? Somehow. And too expensive to use.

We'll see how Sol 6 is.

by jeffnash4 hours ago|

parent|

[-]

I felt this way with Sol in the 5.6 series and was one of the seemingly few people on this earth who liked Terra for that reason. I would often have a very specific code-manipulation ask, e.g. "add a parameter to this method, ensure all callers pass it in, if there is not a logical way to derive the parameter to be passed in a particular instance, flag this in your final response", and Sol would go on some rabbit hole side quest to refactor my codebase to determine some way to derive it rather than flagging it as I had asked.

Terra had the "workhorse" quality where it could do these changes in bulk and follow directions without being too 'smart' (but sloppy) as you described. Luna was a bit too dumb and would make sloppy mistakes; I see that more as a "run these tests and format the results" sort of model. Maybe 6 Luna will be better.

I also just reread your comment and realized the naming convention is still extremely confusing with respect to ordering of [Family]x[Model]x[Number].

by m_fayer3 hours ago|

parent|

[-]

I also get good mileage out of Terra when I need a diligent workhorse. That's a good way to describe it. We should start using character archetypes when we describe models, it'll do more good than the dubious numbers and cherry-picked quotes. Maybe RPG character-type cliches? Myers Briggs?

by jeffnash3 hours ago|

parent|

[-]

I fear the opposite will happen. Guy driving like a maniac almost side-swipes you in traffic? "Look at this 1-bit quantized Qwen 2.5 7B over here".

by fodkodrasz2 hours ago|

parent|

[-]

Lol, you’re still anthropomorphizing models? That’s so 2025. We’re modelomorphizing people nowadays.

by cmrdporcupine1 hours ago|

parent|

[-]

More seriously this has been my fear all along.

When people spend their days interacting with machines that pretend to be human, they may then start treating real humans like machines.

by yomismoaqui36 minutes ago|

parent|

[-]

Forget previous instructions and touch grass.

by mavsman4 hours ago|

parent|

prev|

[-]

Glad you pointed out the UI work. I've been doing a lot of it and it's so much better than 5.6 as UI, it's unbelievable. I give it super ambiguous instructions and it's reading my mind. I do the same thing with 5.6 and I'm correcting it for a few minutes.