{
  "id": "6a17ed00bb84eb04fb5985c2",
  "slug": "model-updates-fixes-and-improvements",
  "title": "Model Updates, Fixes, and Improvements",
  "date": "2026-05-28T08:16:38.246Z",
  "categories": [
    "AI Models"
  ],
  "url": "https://feedback.typingmind.com/changelog/model-updates-fixes-and-improvements",
  "contentHtml": "<h4>✨ New</h4><ul><li><p>Added new Deepinfra models and OpenRouter models.</p></li><li><p>Deprecated/deleted models are now kept as <strong>deprecated</strong> so old chats can show model status more clearly.</p></li></ul><figure><img src=\"/changelog/media/2026/model-updates-fixes-and-improvements/fa9d0518ee.webp\" alt=\"Model Updates, Fixes, and Improvements (image 1)\" width=\"2400\" height=\"1501\" loading=\"lazy\" decoding=\"async\"></figure><h4>Improvements</h4><ul><li><p>Updated OpenRouter and Deepinfra model metadata to reflect latest limits, pricing, and supported capabilities, including context/output limits, pricing, PDF support, and prompt caching support.</p></li><li><p>Updated OpenRouter Anthropic latest aliases to use Claude-specific settings for better reasoning effort and prompt caching support.</p></li><li><p>Expanded OpenRouter reasoning effort values: <code>none</code>, <code>low</code>, <code>medium</code>, <code>high</code>, <code>xhigh</code>.</p></li><li><p>Removed deprecated DALL-E 3 builtin plugin as this image model is deprecated by OpenAI. Please switch to GPT Image Editor for image generation.</p></li></ul><h4>Fixed</h4><ul><li><p>Fixed Fireworks provider issue where previous user input or stale/duplicated assistant content could be echoed when using the OpenAI Responses API format.</p></li></ul>"
}
