Fable 5 is WAY better than Opus 5. It is not even close.
Opus 5 is slow and it is lazy. It stops early, skips steps, and tells you it is done when it is not.
I now launch three subagents on tasks Fable 5 one shots. That is my actual workflow.
I called Opus 5 the best model in the world two weeks ago. I was wrong.
Anthropic dropped it on a Friday out of nowhere. That said everything.
Fable 5.1 is the one I am waiting for.
Fable - Sol loops found another 23% memory reduction in the already very optimized sorted set PR. Now, let's explore radically different approaches compared to the zkiplist+hash approach.
Fable 5 from Anthropic is very very good. The only downside is that in 5 hours of work yesterday, I burned over 100 Euros! I am not sure that is sustainable though!