Fable 5.1 Shows a Shift: More Tool Calls, Shorter Reasoning

Key Info

Fable 5.1 is showing a more action-oriented behavior: 78% of its responses end with a tool call, compared with 51% for Fable 5, and reasoning tokens appear on 33% of requests — up from 27% — but in shorter bursts.

Highlights

• More execution, less talk: 78% of Fable 5.1 responses end in a tool call, versus 51% for Fable 5.
• More frequent, briefer reasoning: 33% of requests include reasoning tokens, up from 27%, but each reasoning burst is shorter.
• Combined with earlier data, median response length has dropped from 603 tokens to 270 tokens — fewer explanations, more doing.