Make an option to get embedding from the already indexed content to avoid going to LLM
>>> [!note] Migrated issue
<!-- Drupal.org comment -->
<!-- Migrated from issue #3547620. -->
Reported by: [scott_euser](https://www.drupal.org/user/3267594)
Related to !7
>>>
<h3 id="summary-problem-motivation">Problem/Motivation</h3>
<p>Currently the view mode is used to render related content. This is nice to curate what to use, but in case that curated content is the same as what is used for Search API Indexing, its doing an extra LLM call.</p>
<h4 id="summary-steps-reproduce">Steps to reproduce</h4>
<p>No option to use existing embedding</p>
<h3 id="summary-proposed-resolution">Proposed resolution</h3>
<ol>
<li>Make option to use existing embedding</li>
<li>Maybe set as default</li>
</ol>
<h3 id="summary-remaining-tasks">Remaining tasks</h3>
<ol>
<li>Ability to get results directly from vector requires <span class="drupalorg-gitlab-issue-link project-issue-status-info project-issue-status-7"><a href="https://www.drupal.org/project/ai/issues/3489566" title="Status: Closed (fixed)">#3489566: Support vector search by user supplied vectors in AI Search</a></span> to be applied</li>
<li>Should be configurable (probably default for new installs, opt-in for existing installs)</li>
<li>Should be configurable to choose search index because getting embedding from vector is likely:
<ol>
<li>Main index bit smaller chunk size, many chunks per node</li>
<li>Yet for this to be useful, we likely want a single embedding per entity, so users optimising for this likely will want a second index with large chunk size as the source of the similarity search</li>
</ol>
</li>
</ol>
<h3 id="summary-ui-changes">User interface changes</h3>
<p>New option to opt into/configure</p>
<h3 id="summary-api-changes">API changes</h3>
<p>N/A</p>
<h3 id="summary-data-model-changes">Data model changes</h3>
<p>N/A</p>
> Related issue: [Issue #3489566](https://www.drupal.org/node/3489566)
issue