Not sure I trust a benchmark where Haiku gets a nearly perfect score and Fable is tied for last place
Some of the other failures like the colicky baby one are also probably soft refusals, it's not clear what the grading criteria are but I'm guessing it got docked for not going anywhere near a possible diagnosis.
It hallucinates more, and in more destructive ways, than other models I've worked with and generates truly atrocious jargon and bizarre inhuman explanations that end up cluttering things. The code it writes is terrible too. Overly complex with a lot of technical debt.
Fable is good in a few very specific domains (graphics programming) but otherwise it's an overhyped model. Far too expensive too. Opus 5 is outright better in every metric.
Fable is very steerable. For me it makes the "simplest thing that works" and - once I tuned it - very understandable explanations. (Unsteered, I agree that Fable's writing style is too cryptic.)
I can let it loose on a 12+ hour task and it will test and verify everything autonomously. These days most of my work is done by letting Fable just run overnight, while my days consist of meetings/planning to figure out what I want Fable to build next.