Hacker News Reader

Comment by mmastrac | Hacker News Reader

I previously posted that DS4Flash was _good_ but not _great_ on two DGX Sparks, but I have to say that GLM-5.3 is pretty amazing. It's been able to tackle all the random hard problems I've thrown at it and it has the intuition that DS4Flash seems to lack.

We're nowhere near a Fable-class model IMO, but things are going to get interesting in this next year.

View on HN

It's early days, but GLM 5.3 Flash is the first local model that feels good enough to me to be a "main" model without debating whether each problem needs to be sent to a stronger model. DS4 Flash is good enough at implementing given a plan, but I wasn't always a fan of what it came up with when asked to plan something.

The good news is it can only get better from here.

> We're nowhere near a Fable-class model IMO, but things are going to get interesting in this next year.

I'm wondering of you could clarify your thoughts on this. I've had a hard time evaluating what Fable-class actually is capable of that sets them (or really it) apart from other models in a very significant way.

Have you tried it out? I doubt there is a general accepted definition. For me it is just more capable of deep reasoning/handling complexity. Still can mess up, still does not count as strong AI - but a level above Opus and co.

Could you give me an example of a similar task you asked Fable and a different model to do where Fable did a better job?

I have a hard time getting models like GLM 5.3 to not perform on my tasks but I might be biased.

Changes in a complex codebase. Opus can do it, but needs more handholding. Making the plan with fable and let opus implement it worked out well.

That's actually not quite what I was asking for.

I assume that's about 5.3 Flash, not full?

Yes, sorry 5.3 flash.

What coding harness are you using? I am trying to take the plunge and wondering which I should use

What quant are you running and tps?

NVFP4 ~20-30tps (MTP + vision, no dflash2).

do you mean GLM 5.3 flash?

Reply on news.ycombinator.com