# The Race to Done: Fable 5.1 vs GPT-6 Astra. Who Wins?

## Executive summary

This video compares the capabilities of Claude Fable 5.1 and GPT-6 Astra by having both models build a native Mac clipboard manager from the same initial prompt. The speaker argues that the true measure of a model's utility is not its initial benchmark performance, but its ability to successfully iterate and refine the build through subsequent, detailed prompting. Astra was preferred due to its speed of iteration, lower token usage, and ease of implementing complex functional changes, such as hotkey adjustments and confirmation messages.

## Key takeaways

- Iteration is the ultimate test of AI utility: The most critical skill is knowing how to refine and improve a model's initial output (the 'second round of prompting'), rather than relying solely on the initial prompt response.
- Model design differences reveal user needs: Comparing two models on the same task (e.g., Fable's narrow list view 'Ledge' vs. Astra's wide card view 'Shelf') helps the user discover design preferences they hadn't consciously decided upon.
- Speed of iteration impacts quality: The speaker found that Astra's ability to process multiple changes (1.0, 1.1, 1.2) in the time it took Fable to complete version 1.0 allowed for more comprehensive refinement and higher quality output.

## Technical details

- Initial Prompt & Goal: Both models were given the prompt: 'Please, build me a native Mac clipboard manager that feels like Apple should have included it. Something that remembers the text, the images, the links I've copied, lets me pull them back with a keyboard shortcut.'
- Model Outputs (Design): Fable built 'Ledge' (a narrow list view sliding in from the right). Astra built 'Shelf' (a wide bar across the bottom with large cards and a preview pane).
- Refinement and Functionality: Key improvements requested and implemented by Astra included: fixing misaligned cards, repositioning the panel via drag-and-drop, correcting interfering key combinations, and adding a visible confirmation message after a successful copy.
- Performance Metrics: The speaker claimed Astra was able to complete the iterative build with fewer tokens and faster speed compared to Fable.

## Practical implications

- When using LLMs for software development, focus on the iterative process (the 'how to get from done') rather than just the initial prompt.
- Use different models for different cognitive tasks: Fable for deep problem thinking/structuring, and Astra for quick execution, drafting, and computer integration.
- The speed of iteration (fast feedback loops) is a critical factor that can outweigh initial model performance metrics.

## Topics

Generative AI, LLM Comparison, Software Development, Mac Development, Prompt Engineering, Shelf & Ledge Apps, Fable 5.1 vs Astra Guide

Source: https://www.youtube.com/watch?v=n5bZHETCiJA