|
21 | 21 |
|
22 | 22 |
|
23 | 23 | <link rel="icon" href="../../assets/favicon.ico"> |
24 | | - <meta name="generator" content="mkdocs-1.6.1, mkdocs-material-9.7.4"> |
| 24 | + <meta name="generator" content="mkdocs-1.6.1, mkdocs-material-9.7.5"> |
25 | 25 |
|
26 | 26 |
|
27 | 27 |
|
|
1096 | 1096 | </span> |
1097 | 1097 | </a> |
1098 | 1098 |
|
| 1099 | +</li> |
| 1100 | + |
| 1101 | + <li class="md-nav__item"> |
| 1102 | + <a href="#endpoint-override" class="md-nav__link"> |
| 1103 | + <span class="md-ellipsis"> |
| 1104 | + |
| 1105 | + Endpoint override |
| 1106 | + |
| 1107 | + </span> |
| 1108 | + </a> |
| 1109 | + |
1099 | 1110 | </li> |
1100 | 1111 |
|
1101 | 1112 | <li class="md-nav__item"> |
|
1541 | 1552 | </span> |
1542 | 1553 | </a> |
1543 | 1554 |
|
| 1555 | +</li> |
| 1556 | + |
| 1557 | + <li class="md-nav__item"> |
| 1558 | + <a href="#endpoint-override" class="md-nav__link"> |
| 1559 | + <span class="md-ellipsis"> |
| 1560 | + |
| 1561 | + Endpoint override |
| 1562 | + |
| 1563 | + </span> |
| 1564 | + </a> |
| 1565 | + |
1544 | 1566 | </li> |
1545 | 1567 |
|
1546 | 1568 | <li class="md-nav__item"> |
@@ -1624,6 +1646,26 @@ <h2 id="modelref-backend-reference">ModelRef (backend reference)</h2> |
1624 | 1646 | </tbody> |
1625 | 1647 | </table> |
1626 | 1648 | <p>The controller that reconciles MaaSModelRef uses <strong>kind</strong> to decide how to resolve the backend and populate <code>status.endpoint</code> and <code>status.phase</code>. Cross-namespace references are supported by specifying <code>modelRef.namespace</code>.</p> |
| 1649 | +<h2 id="endpoint-override">Endpoint override</h2> |
| 1650 | +<p>MaaSModel supports an optional <code>spec.endpointOverride</code> field. When set, the controller uses this value for <code>status.endpoint</code> instead of the auto-discovered endpoint from the backend (LLMInferenceService status, Gateway, or HTTPRoute hostnames).</p> |
| 1651 | +<p>This is useful when: |
| 1652 | +- The controller picks the wrong gateway or hostname for the model endpoint. |
| 1653 | +- Your environment requires a specific URL that differs from what the backend reports. |
| 1654 | +- You need to point the model at a custom proxy or load balancer.</p> |
| 1655 | +<p>Example:</p> |
| 1656 | +<div class="highlight"><pre><span></span><code><a id="__codelineno-0-1" name="__codelineno-0-1" href="#__codelineno-0-1"></a><span class="nt">apiVersion</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">maas.opendatahub.io/v1alpha1</span> |
| 1657 | +<a id="__codelineno-0-2" name="__codelineno-0-2" href="#__codelineno-0-2"></a><span class="nt">kind</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">MaaSModel</span> |
| 1658 | +<a id="__codelineno-0-3" name="__codelineno-0-3" href="#__codelineno-0-3"></a><span class="nt">metadata</span><span class="p">:</span> |
| 1659 | +<a id="__codelineno-0-4" name="__codelineno-0-4" href="#__codelineno-0-4"></a><span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">my-model</span> |
| 1660 | +<a id="__codelineno-0-5" name="__codelineno-0-5" href="#__codelineno-0-5"></a><span class="w"> </span><span class="nt">namespace</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">opendatahub</span> |
| 1661 | +<a id="__codelineno-0-6" name="__codelineno-0-6" href="#__codelineno-0-6"></a><span class="nt">spec</span><span class="p">:</span> |
| 1662 | +<a id="__codelineno-0-7" name="__codelineno-0-7" href="#__codelineno-0-7"></a><span class="w"> </span><span class="nt">modelRef</span><span class="p">:</span> |
| 1663 | +<a id="__codelineno-0-8" name="__codelineno-0-8" href="#__codelineno-0-8"></a><span class="w"> </span><span class="nt">kind</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">LLMInferenceService</span> |
| 1664 | +<a id="__codelineno-0-9" name="__codelineno-0-9" href="#__codelineno-0-9"></a><span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">my-model</span> |
| 1665 | +<a id="__codelineno-0-10" name="__codelineno-0-10" href="#__codelineno-0-10"></a><span class="w"> </span><span class="nt">namespace</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">llm</span> |
| 1666 | +<a id="__codelineno-0-11" name="__codelineno-0-11" href="#__codelineno-0-11"></a><span class="w"> </span><span class="nt">endpointOverride</span><span class="p">:</span><span class="w"> </span><span class="s">"https://correct-hostname.example.com/my-model"</span> |
| 1667 | +</code></pre></div> |
| 1668 | +<p>The controller still validates the backend (HTTPRoute exists, LLMInferenceService is ready, etc.) — the override only affects the final endpoint URL written to <code>status.endpoint</code>. When the field is empty or omitted, the controller uses its normal discovery logic.</p> |
1627 | 1669 | <h2 id="current-behavior">Current behavior</h2> |
1628 | 1670 | <ul> |
1629 | 1671 | <li><strong>Supported kind today:</strong> <code>LLMInferenceService</code> (also accepts the alias <code>llmisvc</code> for backwards compatibility). The MaaS controller reconciles MaaSModelRefs whose <strong>modelRef</strong> points to an LLMInferenceService (by name and optional namespace). It sets <code>status.endpoint</code> from the LLMInferenceService status and <code>status.phase</code> from its readiness.</li> |
@@ -1693,7 +1735,7 @@ <h2 id="summary">Summary</h2> |
1693 | 1735 | <span class="md-icon" title="Last update"> |
1694 | 1736 | <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M21 13.1c-.1 0-.3.1-.4.2l-1 1 2.1 2.1 1-1c.2-.2.2-.6 0-.8l-1.3-1.3c-.1-.1-.2-.2-.4-.2m-1.9 1.8-6.1 6V23h2.1l6.1-6.1zM12.5 7v5.2l4 2.4-1 1L11 13V7zM11 21.9c-5.1-.5-9-4.8-9-9.9C2 6.5 6.5 2 12 2c5.3 0 9.6 4.1 10 9.3-.3-.1-.6-.2-1-.2s-.7.1-1 .2C19.6 7.2 16.2 4 12 4c-4.4 0-8 3.6-8 8 0 4.1 3.1 7.5 7.1 7.9l-.1.2z"/></svg> |
1695 | 1737 | </span> |
1696 | | - <span class="git-revision-date-localized-plugin git-revision-date-localized-plugin-date" title="March 3, 2026 21:33:09 UTC">March 3, 2026</span> |
| 1738 | + <span class="git-revision-date-localized-plugin git-revision-date-localized-plugin-date" title="March 10, 2026 19:12:18 UTC">March 10, 2026</span> |
1697 | 1739 | </span> |
1698 | 1740 |
|
1699 | 1741 |
|
|
0 commit comments