News for hackers. Stuff that matters.

A comment on November 2026 Claude Fable 5.2 Scores 65 on Artificial Analysis, Highest So Far.

Re:November 2026 Claude Fable 5.2 Scores 65 on Artificial An... by GMoromisato on Monday November 09, 2026 @06:56PM (#50434156)
Anecdotal, but I have the model write a podcast script of two reviewers talking about a book, one chapter at a time.

Fable (5.1 and 5.2) consistently write a much shorter script (2.5 pages) compared to Opus 5.5 (4.5 pages).

Fable 5.2: https://claude.ai/share/274a96a8-c8a5-4564-9ca1-e914edffc722

Opus 5.5: https://claude.ai/share/da8ab6bd-0987-47dc-9c26-68ebbfb25781

[ Parent | on HN ]
Re:November 2026 Claude Fable 5.2 Scores 65 on Artificial An... by solarkraft on Monday November 09, 2026 @07:00PM (#50434198)
The reasoning setting is so important. I find that at different levels the solutions differ a lot.
[ Parent | on HN ]
Re:November 2026 Claude Fable 5.2 Scores 65 on Artificial An... by staticman2 on Tuesday November 10, 2026 @09:31AM (#50440934)
Out of curiosity do you know if results would be better, the same, or worse if this was done from a harness?
[ Parent | on HN ]