{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreialt5qb7osob2i6g22rw63m76gnsah5ndvrfdamtqwzylylzxq4xe",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mljhzosfxq52"
  },
  "path": "/t/realtime-regression-in-non-english-production-voice-agents-gpt-realtime-mini-vs-gpt-realtime-mini-2025-10-06/1380643#post_2",
  "publishedAt": "2026-05-10T17:51:39.000Z",
  "site": "https://community.openai.com",
  "tags": [
    "developers.openai.com",
    "Using realtime models | OpenAI API"
  ],
  "textContent": "Hi!\n\nThank you for raising this.\n\nYou are probably already aware that you can still use gpt-realtime-mini-2025-10-06. You do not need to use the undated model slug, which currently points to the December snapshot.\n\nFor production systems, dated model versions are usually the better choice because they help keep behavior more consistent.\n\nSince I do not see a deprecation notice for the older snapshot, using it seems reasonable here.\n\nIn this case, I would suggest staying on the older snapshot for now and waiting for a possible gpt-realtime-mini-2 release, or reviewing the Realtime model best practices here:\n\ndevelopers.openai.com\n\n### Using realtime models | OpenAI API\n\nLearn how to use realtime voice models, including Realtime 2 reasoning, preambles, tool use, and migration from earlier realtime models.",
  "title": "Realtime regression in non-English production voice agents: gpt-realtime-mini vs gpt-realtime-mini-2025-10-06"
}