{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreibutprsuq24sxplzvvjwx3bi464i4zdrlv4m4be7bxum24dvxczrm",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mmlqjpaatgk2"
  },
  "path": "/t/realtime-v2-is-giving-long-responses/1381610#post_4",
  "publishedAt": "2026-05-24T08:39:16.000Z",
  "site": "https://community.openai.com",
  "tags": [
    "conversation item"
  ],
  "textContent": "I am not delivering any products with realtime, so I don’t have particular applications that need refinement where I experience differences in symptom cross model, except for experimentally using them, and seeing more the quality in voices and steerability.\n\nWhat I notice as part of the product envelope, that I can bring to your attention, is the client event to create a conversation item, which can be a mid-session tune up for increased following (one that will be part of the conversation that can be truncated out), and can be textual and a system role message.\n\n> `RealtimeConversationItemSystemMessage`\n>  A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation’s behavior, use instructions, but for smaller updates (e.g. “the user is now asking about a different topic”), use system messages.\n\nSo, you have:\n\n  * tune up gpt-realtime-2 with messaging;\n  * don’t migrate, wait for a model that can fulfill your needs.\n\n",
  "title": "Realtime V2 is giving long responses"
}