diff --git a/web/src/views/settings/SearchSettingsView.tsx b/web/src/views/settings/SearchSettingsView.tsx
index ccf3a5629..f5d80679d 100644
--- a/web/src/views/settings/SearchSettingsView.tsx
+++ b/web/src/views/settings/SearchSettingsView.tsx
@@ -161,10 +161,6 @@ export default function SearchSettingsView({
Semantic Search in Frigate allows you to find tracked objects
within your review items using either the image itself, a
user-defined text description, or an automatically generated one.
- This feature works by creating embeddings — numerical vector
- representations — for both the images and text descriptions of
- your tracked objects. By comparing these embeddings, Frigate
- assesses their similarities to deliver relevant search results.
@@ -182,52 +178,53 @@ export default function SearchSettingsView({
- Re-indexing will reprocess all thumbnails and descriptions (if
- enabled) and apply the embeddings on each startup. Don't forget
- to disable the option after restarting!
+
+
+ Re-indexing will reprocess all thumbnails and descriptions (if
+ enabled) and apply the embeddings on each startup. Don't forget to
+ disable the option after restarting!
+
-
-
+
-
Model Size
-
+
Model Size
+
The size of the model used for semantic search embeddings.
Using small employs a quantized version of the
- model that uses much less RAM and runs faster on CPU with a
- very negligible difference in embedding quality.
+ model that uses less RAM and runs faster on CPU with a very
+ negligible difference in embedding quality.