Benchmarking Fable 5.2 vs Opus 5.5
retained text
Figure I'd post this since it's Opus vs Fable on the same codebase ***as a reviewer*** For context: I only learned recently that what I'm doing is creating a "software factory" for some applications I'm building with shared modules. I tend to agree with the other poster who said Fable 5.2 isn't mind blowing. I'm glad that Anthropic gave us Fable only resets so I didn't feel bad at all testing it. My take: 1. I will continue using Fable like I used to: planning sessions for complicated features and as a reviewer for sensitive work 2. Never orchestrate with it, it's a total waste for that, and it's slow to respond if I need to ask a question (NOTE: reposting this because I accidentally attached the same image twice)
document fetched 2026-11-18 00:44 · canonical https://reddit.com/r/ClaudeCode/comments/1xylf2m/benchmarking_fable_52_vs_opus_55