Summary
Nate Herk puts Fable 5 and Fable 5.1 head-to-head in a real-world build challenge, tasking both agents with creating OpsFlow—a local-first visual automation studio for designing and simulating incident response workflows. Rather than writing code directly, each model acts as an orchestrator, delegating engineering, debugging, and design tasks to subordinate Opus and Sonnet workers according to an identical prompt.
The results reveal striking practical differences: Fable 5 produced a working app for roughly $500 with half a day of runtime, while Fable 5.1 ran for a day and a half at nearly $1,200. Herk walks through both UIs side-by-side, comparing responsiveness, node-based workflow editing, approval handling, light/dark mode support, and zoom behavior. A notable technical divergence emerges: the two versions generated incompatible JSON schemas, meaning workflows built for one cannot be imported into the other.
Beyond aesthetics, the video raises practical questions about multi-agent orchestration strategy and how each model interpreted identical instructions differently—useful context for teams weighing whether upgrading to the latest model version is worth the cost premium in autonomous coding pipelines.
📺 Source: Nate Herk | AI Automation · Published September 03, 2026
🏷️ Format: Comparison
