{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreigevvytgenildx67s332gfhedh7bhxw45h3y7hgzeba3bjwfnibam",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mlwjskl45fe2"
  },
  "path": "/t/why-do-gpt-5-1-and-gpt-5-4-mini-behave-so-differently-in-production-chatbot-use-cases/1380891#post_7",
  "publishedAt": "2026-05-15T23:17:12.000Z",
  "site": "https://community.openai.com",
  "textContent": "I would not rely on the scorecards or generic indexes/benchmarks for production use cases. Use a strong model to build yourself a dataset of a few hundred examples for your specific use case and then benchmark your use case (for quality, speed, cost) with a few models and parameters. You’ll be surprised of what you discover.",
  "title": "Why do gpt-5.1 and gpt-5.4-mini behave so differently in production chatbot use cases?"
}