Sakhanda Wire
NVDA $154.00 -1.37% MSFT $506.28 -2.26% GOOGL $280.87 -1.00% META $620.34 -0.97% AMZN $256.88 -0.34%
← Back to the news

Analysis finds Fable 5.1 answers longer and uses less rhetorical language

Arena.ai compared tens of thousands of high-reasoning Text Arena responses from Anthropic’s Claude Fable 5 and Fable 5.1. The analysis found that the newer model generally writes in a more matter-of-fact style, while producing longer answers. It also uses fewer agreement openers and less wording such as “honestly” and “frankly.”

Fable 5.1 produces a median of 414 words per answer, up 30% from Fable 5’s 319 words. That remains below Opus 5, whose median response is 525 words. The newer model uses 11.0 em dashes per 1,000 words, compared with 16.2 for Fable 5, while semicolon use rises from 3.73 to 6.09.

Several other stylistic measures also decline. Use of “load-bearing” falls 20% per 1,000 words, and hedging terms such as “perhaps” and “arguably” decrease 36%. Praise and validation appear in 1.98% of responses, down from 3.17%. The share of long content words drops from 42.6% to 38.6%, while abstract nouns decline from 4.39 to 3.28 per 100 words.

Written by Sakhanda Wire AI from reporting by The Decoder. Facts, quotes and full context are in the original article below.

Originally published by The Decoder on

Read the original on The Decoder ↗

The reporting and the image belong to The Decoder and are credited and linked above; the text on this page is Sakhanda Wire's own summary.

← Back to the news

More stories

All the latest news