[1.16] Add ACORN and Inline Storage docs (#1990)

* Add ACORN doc * Add Inline Storage doc * Use consistent version gate marker * Update --------- Co-authored-by: Tim Visée <tim@visee.me>
2026-10-04 10:28:29 +02:00 · 2025-11-17 15:16:51 +01:00
parent 4bf27cb600
commit 69b90aadce
19 changed files with 363 additions and 8 deletions
@@ -0,0 +1,4 @@
+When creating a collection with inline storage enabled, the HNSW index stores copies of both the original vectors and quantized vectors within the index file itself.
+This reduces random disk seeks during search, trading disk space for improved search speed.
+The `inline_storage` option requires quantization to be enabled and does not support multi-vectors.
+Set `hnsw_config.inline_storage` to `true` and configure quantization to use this feature.
@@ -0,0 +1,16 @@
+```csharp
+using Qdrant.Client;
+using Qdrant.Client.Grpc;
+
+var client = new QdrantClient("localhost", 6334);
+
+await client.CreateCollectionAsync(
+	collectionName: "{collection_name}",
+	vectorsConfig: new VectorParams { Size = 768, Distance = Distance.Cosine, OnDisk = true },
+	quantizationConfig: new QuantizationConfig
+	{
+		Binary = new BinaryQuantization { AlwaysRam = false }
+	},
+	hnswConfig: new HnswConfigDiff { OnDisk = true, InlineStorage = true }
+);
+```
@@ -0,0 +1,30 @@
+```go
+import (
+	"context"
+
+	"github.com/qdrant/go-client/qdrant"
+)
+
+client, err := qdrant.NewClient(&qdrant.Config{
+	Host: "localhost",
+	Port: 6334,
+})
+
+client.CreateCollection(context.Background(), &qdrant.CreateCollection{
+	CollectionName: "{collection_name}",
+	VectorsConfig: qdrant.NewVectorsConfig(&qdrant.VectorParams{
+		Size:     768,
+		Distance: qdrant.Distance_Cosine,
+		OnDisk:   qdrant.PtrOf(true),
+	}),
+	QuantizationConfig: qdrant.NewQuantizationBinary(
+		&qdrant.BinaryQuantization{
+			AlwaysRam: qdrant.PtrOf(false),
+		},
+	),
+	HnswConfig: &qdrant.HnswConfigDiff{
+		OnDisk:        qdrant.PtrOf(true),
+		InlineStorage: qdrant.PtrOf(true),
+	},
+})
+```
@@ -0,0 +1,19 @@
+```http
+PUT /collections/{collection_name}
+{
+    "vectors": {
+      "size": 768,
+      "distance": "Cosine",
+      "on_disk": true
+    },
+    "quantization_config": {
+        "binary": {
+            "always_ram": false
+        }
+    },
+    "hnsw_config": {
+        "on_disk": true,
+        "inline_storage": true
+    }
+}
+```
@@ -0,0 +1,35 @@
+```java
+import io.qdrant.client.QdrantClient;
+import io.qdrant.client.QdrantGrpcClient;
+import io.qdrant.client.grpc.Collections.BinaryQuantization;
+import io.qdrant.client.grpc.Collections.CreateCollection;
+import io.qdrant.client.grpc.Collections.Distance;
+import io.qdrant.client.grpc.Collections.HnswConfigDiff;
+import io.qdrant.client.grpc.Collections.QuantizationConfig;
+import io.qdrant.client.grpc.Collections.VectorParams;
+import io.qdrant.client.grpc.Collections.VectorsConfig;
+
+QdrantClient client =
+    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());
+
+client
+    .createCollectionAsync(
+        CreateCollection.newBuilder()
+            .setCollectionName("{collection_name}")
+            .setVectorsConfig(
+                VectorsConfig.newBuilder()
+                    .setParams(
+                        VectorParams.newBuilder()
+                            .setSize(768)
+                            .setDistance(Distance.Cosine)
+                            .setOnDisk(true)
+                            .build())
+                    .build())
+            .setQuantizationConfig(
+                QuantizationConfig.newBuilder()
+                    .setBinary(BinaryQuantization.newBuilder().setAlwaysRam(false).build())
+                    .build())
+            .setHnswConfig(HnswConfigDiff.newBuilder().setOnDisk(true).setInlineStorage(true).build())
+            .build())
+    .get();
+```
@@ -0,0 +1,16 @@
+```python
+from qdrant_client import QdrantClient, models
+
+client = QdrantClient(url="http://localhost:6333")
+
+client.create_collection(
+    collection_name="{collection_name}",
+    vectors_config=models.VectorParams(
+        size=768, distance=models.Distance.COSINE, on_disk=True
+    ),
+    quantization_config=models.BinaryQuantization(
+        binary=models.BinaryQuantizationConfig(always_ram=False),
+    ),
+    hnsw_config=models.HnswConfigDiff(on_disk=True, inline_storage=True),
+)
+```
@@ -0,0 +1,21 @@
+```rust
+use qdrant_client::qdrant::{
+    BinaryQuantizationBuilder, CreateCollectionBuilder, Distance, HnswConfigDiffBuilder,
+    VectorParamsBuilder,
+};
+use qdrant_client::Qdrant;
+
+let client = Qdrant::from_url("http://localhost:6334").build()?;
+
+client
+    .create_collection(
+        CreateCollectionBuilder::new("{collection_name}")
+            .vectors_config(VectorParamsBuilder::new(768, Distance::Cosine).on_disk(true))
+            .quantization_config(BinaryQuantizationBuilder::new(false))
+            .hnsw_config(
+                HnswConfigDiffBuilder::default()
+                    .on_disk(true)
+                    .inline_storage(true),
+            ),
+    )
+    .await?;
@@ -0,0 +1,22 @@
+```typescript
+import { QdrantClient } from "@qdrant/js-client-rest";
+
+const client = new QdrantClient({ host: "localhost", port: 6333 });
+
+client.createCollection("{collection_name}", {
+  vectors: {
+    size: 768,
+    distance: "Cosine",
+    on_disk: true,
+  },
+  quantization_config: {
+    binary: {
+      always_ram: false,
+    },
+  },
+  hnsw_config: {
+    on_disk: true,
+    inline_storage: true,
+  },
+});
+```
@@ -0,0 +1,6 @@
+This code snippet demonstrates how to enable ACORN for HNSW search.
+ACORN improves search recall for searches with multiple low-selectivity payload filters, at the cost of reduced performance.
+The `enable` parameter activates ACORN based on filter selectivity.
+The `max_selectivity` parameter controls the threshold - if estimated filter selectivity is higher than this value, ACORN will not be used.
+Selectivity is estimated as the ratio of points satisfying the filters to total points.
+Values range from 0.0 (never use ACORN) to 1.0 (always use ACORN), with a default of 0.4.
@@ -0,0 +1,20 @@
+```csharp
+using Qdrant.Client;
+using Qdrant.Client.Grpc;
+
+var client = new QdrantClient("localhost", 6334);
+
+await client.QueryAsync(
+	collectionName: "{collection_name}",
+	query: new float[] { 0.2f, 0.1f, 0.9f, 0.7f },
+	searchParams: new SearchParams
+	{
+		Acorn = new AcornSearchParams
+		{
+			Enable = true,
+			MaxSelectivity = 0.4
+		}
+	},
+	limit: 10
+);
+```
@@ -0,0 +1,23 @@
+```go
+import (
+	"context"
+
+	"github.com/qdrant/go-client/qdrant"
+)
+
+client, err := qdrant.NewClient(&qdrant.Config{
+	Host: "localhost",
+	Port: 6334,
+})
+
+client.Query(context.Background(), &qdrant.QueryPoints{
+	CollectionName: "{collection_name}",
+	Query:          qdrant.NewQuery(0.2, 0.1, 0.9, 0.7),
+	Params: &qdrant.SearchParams{
+		Acorn: &qdrant.AcornSearchParams{
+			Enable:          qdrant.PtrOf(true),
+			MaxSelectivity:  qdrant.PtrOf(0.4),
+		},
+	},
+})
+```
@@ -0,0 +1,13 @@
+```http
+POST /collections/{collection_name}/points/query
+{
+    "query": [0.2, 0.1, 0.9, 0.7],
+    "params": {
+        "acorn": {
+            "enable": true,
+            "max_selectivity": 0.4
+        }
+    },
+    "limit": 10
+}
+```
@@ -0,0 +1,28 @@
+```java
+import io.qdrant.client.QdrantClient;
+import io.qdrant.client.QdrantGrpcClient;
+import io.qdrant.client.grpc.Points.AcornSearchParams;
+import io.qdrant.client.grpc.Points.QueryPoints;
+import io.qdrant.client.grpc.Points.SearchParams;
+
+import static io.qdrant.client.QueryFactory.nearest;
+
+QdrantClient client =
+    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());
+
+client.queryAsync(
+        QueryPoints.newBuilder()
+                .setCollectionName("{collection_name}")
+                .setQuery(nearest(0.2f, 0.1f, 0.9f, 0.7f))
+                .setParams(
+                        SearchParams.newBuilder()
+                                .setAcorn(
+                                        AcornSearchParams.newBuilder()
+                                                .setEnable(true)
+                                                .setMaxSelectivity(0.4)
+                                                .build())
+                                .build())
+                .setLimit(10)
+                .build())
+        .get();
+```
@@ -0,0 +1,17 @@
+```python
+from qdrant_client import QdrantClient, models
+
+client = QdrantClient(url="http://localhost:6333")
+
+client.query_points(
+    collection_name="{collection_name}",
+    query=[0.2, 0.1, 0.9, 0.7],
+    search_params=models.SearchParams(
+        acorn=models.AcornSearchParams(
+            enable=True,
+            max_selectivity=0.4,
+        )
+    ),
+    limit=10,
+)
+```
@@ -0,0 +1,22 @@
+```rust
+use qdrant_client::qdrant::{
+    AcornSearchParamsBuilder, QueryPointsBuilder, SearchParamsBuilder,
+};
+use qdrant_client::Qdrant;
+
+let client = Qdrant::from_url("http://localhost:6334").build()?;
+
+client
+    .query(
+        QueryPointsBuilder::new("{collection_name}")
+            .query(vec![0.2, 0.1, 0.9, 0.7])
+            .limit(10)
+            .params(
+                SearchParamsBuilder::default().acorn(
+                    AcornSearchParamsBuilder::new(true)
+                        .max_selectivity(0.4),
+                ),
+            ),
+    )
+    .await?;
+```
@@ -0,0 +1,16 @@
+```typescript
+import { QdrantClient } from "@qdrant/js-client-rest";
+
+const client = new QdrantClient({ host: "localhost", port: 6333 });
+
+client.query("{collection_name}", {
+    query: [0.2, 0.1, 0.9, 0.7],
+    params: {
+        acorn: {
+            enable: true,
+            max_selectivity: 0.4,
+        },
+    },
+    limit: 10,
+});
+```