{
  "id": "6a9a5d53d9f715f753d990db",
  "slug": "multi-model-response-improvements",
  "title": "Multi-Model Response Improvements",
  "date": "2026-09-04T06:31:26.603Z",
  "categories": [
    "Improved",
    "AI Models"
  ],
  "url": "https://feedback.typingmind.com/changelog/multi-model-response-improvements",
  "contentHtml": "<p>We’ve improved the multi-model experience:</p><ul><li><p>Support stopping and regenerating messages in multi-model responses</p></li></ul><figure><img src=\"/changelog/media/2026/multi-model-response-improvements/460025fb0c.webp\" alt=\"Multi-Model Response Improvements (image 1)\" width=\"2400\" height=\"1580\" loading=\"lazy\" decoding=\"async\"></figure><figure><img src=\"/changelog/media/2026/multi-model-response-improvements/b3d564dfcf.webp\" alt=\"Multi-Model Response Improvements (image 2)\" width=\"2400\" height=\"1508\" loading=\"lazy\" decoding=\"async\"></figure><ul><li><p>Support pasting files and large text as file attachments when replying in multi-model threads.</p></li></ul><figure><img src=\"/changelog/media/2026/multi-model-response-improvements/0de174738f.webp\" alt=\"Multi-Model Response Improvements (image 3)\" width=\"2400\" height=\"1560\" loading=\"lazy\" decoding=\"async\"></figure><ul><li><p>Show statistics such as token usage and estimated cost for each model.</p></li></ul><figure><img src=\"/changelog/media/2026/multi-model-response-improvements/c0fe84c037.webp\" alt=\"Multi-Model Response Improvements (image 4)\" width=\"2400\" height=\"1533\" loading=\"lazy\" decoding=\"async\"></figure><ul><li><p>Fork a chat directly from a multi-model response message.</p></li></ul><figure><img src=\"/changelog/media/2026/multi-model-response-improvements/5af5caf30e.webp\" alt=\"Multi-Model Response Improvements (image 5)\" width=\"2400\" height=\"1580\" loading=\"lazy\" decoding=\"async\"></figure><ul><li><p>When finalizing a multi-model response, you can now click <strong>Edit finalize instruction</strong> to customize the prompt used for that specific finalize request before picking the model to generate it.</p></li></ul><figure><img src=\"/changelog/media/2026/multi-model-response-improvements/d78dabfd96.webp\" alt=\"Multi-Model Response Improvements (image 6)\" width=\"2400\" height=\"1580\" loading=\"lazy\" decoding=\"async\"></figure><p><strong>Note:</strong> The default finalize instruction can now be customized in <strong>Settings → Advanced Settings → Internal Prompts</strong> under a new \"Finalize multi-model responses\" section (with a \"Reset to default\" option), synced across your devices via cloud sync.</p><figure><img src=\"/changelog/media/2026/multi-model-response-improvements/4bbedf7325.webp\" alt=\"Multi-Model Response Improvements (image 7)\" width=\"2400\" height=\"1580\" loading=\"lazy\" decoding=\"async\"></figure>"
}