Add MRL to Inference docs

qdrant-landing/content/documentation/headless/snippets/inference/mrl-multi-stage/_description.md

@@ -0,0 +1 @@
+This code snippet illustrates how to use smaller vectors for the initial prefetching of candidates from a large collection, followed by re-scoring with the original-sized vectors to improve accuracy, combined with inference. For the smaller vector, it employs Matryoshka Representation Learning (MRL) to reduce the dimensionality of embeddings by specifying the `mrl` parameter in the `options` object.

qdrant-landing/content/documentation/headless/snippets/inference/mrl-multi-stage/http.md

@@ -0,0 +1,26 @@
+```http
+POST /collections/{collection_name}/points/query
+{
+  "prefetch": {
+    "query": {
+      "text": "How to bake cookies?",
+      "model": "openai/text-embedding-3-small",
+      "options": {
+        "openai-api-key": "<YOUR_OPENAI_API_KEY>",
+        "mrl": 64
+      }
+    },
+    "using": "small",
+    "limit": 1000
+  },
+  "query": {
+    "text": "How to bake cookies?",
+    "model": "openai/text-embedding-3-small",
+    "options": {
+      "openai-api-key": "<YOUR_OPENAI_API_KEY>"
+    }
+  },
+  "using": "large",
+  "limit": 10
+}
+```

qdrant-landing/content/documentation/headless/snippets/inference/mrl/_description.md

@@ -0,0 +1 @@
+This code snippet illustrates how to reduce the dimensionality of embeddings using Matryoshka Representation Learning (MRL) when using inference. It demonstrates how to insert a point into a Qdrant collection with a reduced-size vector by specifying the `mrl` parameter in the `options` object.

qdrant-landing/content/documentation/headless/snippets/inference/mrl/http.md

@@ -0,0 +1,20 @@
+```http
+PUT /collections/{collection_name}/points?wait=true
+{
+  "points": [
+    {
+      "id": 1,
+      "vector": {
+        "small": {
+          "text": "Recipe for baking chocolate chip cookies",
+          "model": "openai/text-embedding-3-small",
+          "options": {
+            "openai-api-key": "<YOUR_OPENAI_API_KEY>",
+            "mrl": 64
+          }
+        }
+      }
+    }
+  ]
+}
+```