new | show | ask | jobs Github

Loquebantur an hour ago

What you're looking at there is the cost/performance in inference. The "frontier".

An economically successful step would be to improve training cost/frontier advance.

They essentially failed at that, assuming they kept their training costs for the model step at least constant.

▲

XenophileJKO 10 minutes ago | parent | next [-]

This model is on the cost performance frontier (Pareto line) from medium -> max.

If you leave xhigh and max off.. you still get better performance than Opus 5.5 max

Opus 5.5 Max - score 59 - cost per task - $10.04

Fable 5.2 High - score 62 - cost per task - $9.08

▲

lukeschlather 27 minutes ago | parent | prev | next [-]

The thing is max reasoning is most effective at things which are least interesting (the ones where the answers can be validated by a computer.) I've found for a lot of tasks like image recognition, low reasoning can even be better quality than max.

▲

rsalus an hour ago | parent | prev [-]

isn't that what Opus is for? this is the "strongest" model, not the most economical

▲

breakwaterlabs 21 minutes ago | parent [-]

If you believe in things like Artificial Analysis, GPT-6.1-Sol is by far more economical than Opus 5.5, even before the alleged (observed) nerf last week: https://artificialanalysis.ai/?models=gpt-6-astra-high%2Cgpt...

Notable takeaways:

- Sol-low - cost-competitive with Haiku, while outperforming Sonnet on medium

- Sol-High - Competes with Astra medium and beats Opus 5.5-medium, at 20% the cost

- Sol-Max - Beats Opus 5.5 high, Astra high, and competes with Opus xhigh, at cheaper than Sonnet and a fraction of the Opus cost

And when you look at time spent, With the exception of Sol-max you get better results in less time for less money. My own experience is that this is true-- it is difficult to call Opus strictly better than Sol, but Sol gets where it's going far more efficiently.

It's rather hard to look at opus-- or any of anthropic's models-- and say "economical".

	▲	lillecarl 16 minutes ago \| parent [-]
		Their subscription plans are quite economical, API pricing is __absolutely insane__ on the other hand.