Programmer tests Claude vs GPT vs Grok: I'm surprised!
Description: The Claude Fable delegation strategy. Save to your CLAUDE.md: https://gist.github.com/bholmesdev/bdbb473f85fcce74cd1cee5e6da0ee9d
Note: This assumes you have the Codex plugin installed, but you can tweak to say "use the Codex CLI."
0:00 Intro
0:32 Interactive agents vs. cloud automations
1:02 Scoped vs. unscoped work
2:00 Cloud automation examples
2:34 The three metrics: cost, intelligence, taste
3:31 Claude Fable
5:52 Where Fable's taste shines
8:41 GPT 5.6
11:59 Grok 4.5
15:31 Matching models to task types
18:54 Dark horse: frontier model delegation
21:37 Wrap up Close description
Note: This assumes you have the Codex plugin installed, but you can tweak to say "use the Codex CLI."
0:00 Intro
0:32 Interactive agents vs. cloud automations
1:02 Scoped vs. unscoped work
2:00 Cloud automation examples
2:34 The three metrics: cost, intelligence, taste
3:31 Claude Fable
5:52 Where Fable's taste shines
8:41 GPT 5.6
11:59 Grok 4.5
15:31 Matching models to task types
18:54 Dark horse: frontier model delegation
21:37 Wrap up Close description
