OpenAI Codex vs Lovable: what developers actually say
Based on 2,042 opinions on Hacker News from developers who discussed these two, most of them first-hand users.
OpenAI Codex comes out ahead overall on much larger evidence: 2,013 opinions, 1,708 firsthand accounts and 69% approval, against 29 opinions, 10 firsthand accounts and 45% approval for Lovable. Codex's strongest aspect counts are code quality at 343 positive and 116 negative, autonomy at 169 and 46, and speed at 96 and 33; reliability is its weak spot at 112 positive and 127 negative. Usage limits and UX draw complaints, and its 90-day approval is 63% versus 46% for Lovable; only three comments stated a direct preference, two for Lovable. Lovable is the better pick for quick prototypes and non-technical builders, especially CRUD apps with third-party integrations, but its 29 opinions and 10 firsthand accounts are thin, and code quality is 1 positive against 8 negative.
When developers explicitly said they prefer one or switched: 1 picked OpenAI Codex, 2 picked Lovable.
Choose OpenAI Codex if
- Want a terminal agent with deep firsthand HN coverage
- Value strong code-quality and autonomy sentiment
- Can tolerate usage-limit and reliability complaints
Choose Lovable if
- Need a working prototype fast
- Are non-technical or building CRUD apps
- Want deployment, payments and analytics handled in one flow
Where each one wins and loses
Why developers switched
Picked OpenAI Codex
Picked Lovable
Summaries are paraphrased. Each links to the original comment.