GPT‑5.6 Sol outpaces Claude Fable on PRDs, prototypes and browser automation
Lenny’s benchmark shows GPT‑5.6 Sol beats Claude Fable across product‑focused tasks—PRD drafting, rapid prototyping, and agentic voice‑driven browser automation. The margin is large enough to shift model choice for PMs who need speed and reliability, while Sonnet 5 still leads on pure voice tasks.