[8.x] Adding support for binary embedding type to Cohere service embedding type (#120751)

docs/changelog/120751.yaml

@@ -0,0 +1,5 @@
+pr: 120751
+summary: Adding support for binary embedding type to Cohere service embedding type
+area: Machine Learning
+type: enhancement
+issues: []

docs/reference/inference/service-cohere.asciidoc

@@ -62,7 +62,7 @@ include::inference-shared.asciidoc[tag=chunking-settings-strategy]
 
 `service`::
 (Required, string)
-The type of service supported for the specified task type. In this case, 
+The type of service supported for the specified task type. In this case,
 `cohere`.
 
 `service_settings`::
@@ -127,6 +127,8 @@ Valid values are:
 * `byte`: use it for signed int8 embeddings (this is a synonym of `int8`).
 * `float`: use it for the default float embeddings.
 * `int8`: use it for signed int8 embeddings.
+* `binary`: use it for binary embeddings, which are encoded as bytes with signed int8 precision.
+* `bit`: use it for binary embeddings, which are encoded as bytes with signed int8 precision (this is a synonym of `binary`).
 
 `model_id`:::
 (Optional, string)
@@ -228,4 +230,4 @@ PUT _inference/rerank/cohere-rerank
 // TEST[skip:TBD]
 
 For more examples, also review the
-https://docs.cohere.com/docs/elasticsearch-and-cohere#rerank-search-results-with-cohere-and-elasticsearch[Cohere documentation].
+https://docs.cohere.com/docs/elasticsearch-and-cohere#rerank-search-results-with-cohere-and-elasticsearch[Cohere documentation].

server/src/main/java/org/elasticsearch/TransportVersions.java

@@ -178,6 +178,7 @@ static TransportVersion def(int id) {
     public static final TransportVersion TIMEOUT_GET_PARAM_FOR_RESOLVE_CLUSTER = def(8_838_0_00);
     public static final TransportVersion INFERENCE_REQUEST_ADAPTIVE_RATE_LIMITING = def(8_839_0_00);
     public static final TransportVersion ML_INFERENCE_IBM_WATSONX_RERANK_ADDED = def(8_840_0_00);
+    public static final TransportVersion COHERE_BIT_EMBEDDING_TYPE_SUPPORT_ADDED_BACKPORT_8_X = def(8_840_0_01);
 
     /*
      * STOP! READ THIS FIRST! No, really,

x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/inference/results/InferenceByteEmbedding.java

@@ -0,0 +1,95 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License
+ * 2.0; you may not use this file except in compliance with the Elastic License
+ * 2.0.
+ *
+ * this file was contributed to by a generative AI
+ */
+
+package org.elasticsearch.xpack.core.inference.results;
+
+import org.elasticsearch.common.Strings;
+import org.elasticsearch.common.io.stream.StreamInput;
+import org.elasticsearch.common.io.stream.StreamOutput;
+import org.elasticsearch.common.io.stream.Writeable;
+import org.elasticsearch.xcontent.ToXContentObject;
+import org.elasticsearch.xcontent.XContentBuilder;
+
+import java.io.IOException;
+import java.util.Arrays;
+import java.util.List;
+
+public record InferenceByteEmbedding(byte[] values) implements Writeable, ToXContentObject, EmbeddingInt {
+    public static final String EMBEDDING = "embedding";
+
+    public InferenceByteEmbedding(StreamInput in) throws IOException {
+        this(in.readByteArray());
+    }
+
+    @Override
+    public void writeTo(StreamOutput out) throws IOException {
+        out.writeByteArray(values);
+    }
+
+    public static InferenceByteEmbedding of(List<Byte> embeddingValuesList) {
+        byte[] embeddingValues = new byte[embeddingValuesList.size()];
+        for (int i = 0; i < embeddingValuesList.size(); i++) {
+            embeddingValues[i] = embeddingValuesList.get(i);
+        }
+        return new InferenceByteEmbedding(embeddingValues);
+    }
+
+    @Override
+    public XContentBuilder toXContent(XContentBuilder builder, Params params) throws IOException {
+        builder.startObject();
+
+        builder.startArray(EMBEDDING);
+        for (byte value : values) {
+            builder.value(value);
+        }
+        builder.endArray();
+
+        builder.endObject();
+        return builder;
+    }
+
+    @Override
+    public String toString() {
+        return Strings.toString(this);
+    }
+
+    float[] toFloatArray() {
+        float[] floatArray = new float[values.length];
+        for (int i = 0; i < values.length; i++) {
+            floatArray[i] = ((Byte) values[i]).floatValue();
+        }
+        return floatArray;
+    }
+
+    double[] toDoubleArray() {
+        double[] doubleArray = new double[values.length];
+        for (int i = 0; i < values.length; i++) {
+            doubleArray[i] = ((Byte) values[i]).doubleValue();
+        }
+        return doubleArray;
+    }
+
+    @Override
+    public int getSize() {
+        return values().length;
+    }
+
+    @Override
+    public boolean equals(Object o) {
+        if (this == o) return true;
+        if (o == null || getClass() != o.getClass()) return false;
+        InferenceByteEmbedding embedding = (InferenceByteEmbedding) o;
+        return Arrays.equals(values, embedding.values);
+    }
+
+    @Override
+    public int hashCode() {
+        return Arrays.hashCode(values);
+    }
+}

x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/inference/results/InferenceTextEmbeddingBitResults.java

@@ -0,0 +1,109 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License
+ * 2.0; you may not use this file except in compliance with the Elastic License
+ * 2.0.
+ *
+ * this file was contributed to by a generative AI
+ */
+
+package org.elasticsearch.xpack.core.inference.results;
+
+import org.elasticsearch.common.io.stream.StreamInput;
+import org.elasticsearch.common.io.stream.StreamOutput;
+import org.elasticsearch.common.xcontent.ChunkedToXContent;
+import org.elasticsearch.inference.InferenceResults;
+import org.elasticsearch.inference.InferenceServiceResults;
+import org.elasticsearch.xcontent.ToXContent;
+import org.elasticsearch.xpack.core.ml.inference.results.MlTextEmbeddingResults;
+
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.Iterator;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Objects;
+
+/**
+ * Writes a text embedding result in the follow json format
+ * {
+ *     "text_embedding_bytes": [
+ *         {
+ *             "embedding": [
+ *                 23
+ *             ]
+ *         },
+ *         {
+ *             "embedding": [
+ *                 -23
+ *             ]
+ *         }
+ *     ]
+ * }
+ */
+public record InferenceTextEmbeddingBitResults(List<InferenceByteEmbedding> embeddings) implements InferenceServiceResults, TextEmbedding {
+    public static final String NAME = "text_embedding_service_bit_results";
+    public static final String TEXT_EMBEDDING_BITS = "text_embedding_bits";
+
+    public InferenceTextEmbeddingBitResults(StreamInput in) throws IOException {
+        this(in.readCollectionAsList(InferenceByteEmbedding::new));
+    }
+
+    @Override
+    public int getFirstEmbeddingSize() {
+        return TextEmbeddingUtils.getFirstEmbeddingSize(new ArrayList<>(embeddings));
+    }
+
+    @Override
+    public Iterator<? extends ToXContent> toXContentChunked(ToXContent.Params params) {
+        return ChunkedToXContent.builder(params).array(TEXT_EMBEDDING_BITS, embeddings.iterator());
+    }
+
+    @Override
+    public void writeTo(StreamOutput out) throws IOException {
+        out.writeCollection(embeddings);
+    }
+
+    @Override
+    public String getWriteableName() {
+        return NAME;
+    }
+
+    @Override
+    public List<? extends InferenceResults> transformToCoordinationFormat() {
+        return embeddings.stream()
+            .map(embedding -> new MlTextEmbeddingResults(TEXT_EMBEDDING_BITS, embedding.toDoubleArray(), false))
+            .toList();
+    }
+
+    @Override
+    @SuppressWarnings("deprecation")
+    public List<? extends InferenceResults> transformToLegacyFormat() {
+        var legacyEmbedding = new LegacyTextEmbeddingResults(
+            embeddings.stream().map(embedding -> new LegacyTextEmbeddingResults.Embedding(embedding.toFloatArray())).toList()
+        );
+
+        return List.of(legacyEmbedding);
+    }
+
+    public Map<String, Object> asMap() {
+        Map<String, Object> map = new LinkedHashMap<>();
+        map.put(TEXT_EMBEDDING_BITS, embeddings);
+
+        return map;
+    }
+
+    @Override
+    public boolean equals(Object o) {
+        if (this == o) return true;
+        if (o == null || getClass() != o.getClass()) return false;
+        InferenceTextEmbeddingBitResults that = (InferenceTextEmbeddingBitResults) o;
+        return Objects.equals(embeddings, that.embeddings);
+    }
+
+    @Override
+    public int hashCode() {
+        return Objects.hash(embeddings);
+    }
+}

x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/inference/results/InferenceTextEmbeddingByteResults.java

@@ -9,21 +9,16 @@
 
 package org.elasticsearch.xpack.core.inference.results;
 
-import org.elasticsearch.common.Strings;
 import org.elasticsearch.common.io.stream.StreamInput;
 import org.elasticsearch.common.io.stream.StreamOutput;
-import org.elasticsearch.common.io.stream.Writeable;
 import org.elasticsearch.common.xcontent.ChunkedToXContent;
 import org.elasticsearch.inference.InferenceResults;
 import org.elasticsearch.inference.InferenceServiceResults;
 import org.elasticsearch.xcontent.ToXContent;
-import org.elasticsearch.xcontent.ToXContentObject;
-import org.elasticsearch.xcontent.XContentBuilder;
 import org.elasticsearch.xpack.core.ml.inference.results.MlTextEmbeddingResults;
 
 import java.io.IOException;
 import java.util.ArrayList;
-import java.util.Arrays;
 import java.util.Iterator;
 import java.util.LinkedHashMap;
 import java.util.List;
@@ -33,7 +28,7 @@
 /**
  * Writes a text embedding result in the follow json format
  * {
- *     "text_embedding": [
+ *     "text_embedding_bytes": [
  *         {
  *             "embedding": [
  *                 23
@@ -111,78 +106,4 @@ public boolean equals(Object o) {
     public int hashCode() {
         return Objects.hash(embeddings);
     }
-
-    public record InferenceByteEmbedding(byte[] values) implements Writeable, ToXContentObject, EmbeddingInt {
-        public static final String EMBEDDING = "embedding";
-
-        public InferenceByteEmbedding(StreamInput in) throws IOException {
-            this(in.readByteArray());
-        }
-
-        @Override
-        public void writeTo(StreamOutput out) throws IOException {
-            out.writeByteArray(values);
-        }
-
-        public static InferenceByteEmbedding of(List<Byte> embeddingValuesList) {
-            byte[] embeddingValues = new byte[embeddingValuesList.size()];
-            for (int i = 0; i < embeddingValuesList.size(); i++) {
-                embeddingValues[i] = embeddingValuesList.get(i);
-            }
-            return new InferenceByteEmbedding(embeddingValues);
-        }
-
-        @Override
-        public XContentBuilder toXContent(XContentBuilder builder, Params params) throws IOException {
-            builder.startObject();
-
-            builder.startArray(EMBEDDING);
-            for (byte value : values) {
-                builder.value(value);
-            }
-            builder.endArray();
-
-            builder.endObject();
-            return builder;
-        }
-
-        @Override
-        public String toString() {
-            return Strings.toString(this);
-        }
-
-        private float[] toFloatArray() {
-            float[] floatArray = new float[values.length];
-            for (int i = 0; i < values.length; i++) {
-                floatArray[i] = ((Byte) values[i]).floatValue();
-            }
-            return floatArray;
-        }
-
-        private double[] toDoubleArray() {
-            double[] doubleArray = new double[values.length];
-            for (int i = 0; i < values.length; i++) {
-                doubleArray[i] = ((Byte) values[i]).floatValue();
-            }
-            return doubleArray;
-        }
-
-        @Override
-        public int getSize() {
-            return values().length;
-        }
-
-        @Override
-        public boolean equals(Object o) {
-            if (this == o) return true;
-            if (o == null || getClass() != o.getClass()) return false;
-            InferenceByteEmbedding embedding = (InferenceByteEmbedding) o;
-            return Arrays.equals(values, embedding.values);
-        }
-
-        @Override
-        public int hashCode() {
-            return Arrays.hashCode(values);
-        }
-    }
 }

x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/chunking/EmbeddingRequestChunker.java

@@ -19,6 +19,7 @@
 import org.elasticsearch.xpack.core.inference.results.ChunkedInferenceEmbeddingFloat;
 import org.elasticsearch.xpack.core.inference.results.ChunkedInferenceEmbeddingSparse;
 import org.elasticsearch.xpack.core.inference.results.ChunkedInferenceError;
+import org.elasticsearch.xpack.core.inference.results.InferenceByteEmbedding;
 import org.elasticsearch.xpack.core.inference.results.InferenceTextEmbeddingByteResults;
 import org.elasticsearch.xpack.core.inference.results.InferenceTextEmbeddingFloatResults;
 import org.elasticsearch.xpack.core.inference.results.SparseEmbeddingResults;
@@ -69,7 +70,7 @@ public static EmbeddingType fromDenseVectorElementType(DenseVectorFieldMapper.El
 
     private List<ChunkOffsetsAndInput> chunkedOffsets;
     private List<AtomicArray<List<InferenceTextEmbeddingFloatResults.InferenceFloatEmbedding>>> floatResults;
-    private List<AtomicArray<List<InferenceTextEmbeddingByteResults.InferenceByteEmbedding>>> byteResults;
+    private List<AtomicArray<List<InferenceByteEmbedding>>> byteResults;
     private List<AtomicArray<List<SparseEmbeddingResults.Embedding>>> sparseResults;
     private AtomicArray<Exception> errors;
     private ActionListener<List<ChunkedInference>> finalListener;
@@ -389,9 +390,9 @@ private ChunkedInferenceEmbeddingFloat mergeFloatResultsWithInputs(
 
     private ChunkedInferenceEmbeddingByte mergeByteResultsWithInputs(
         ChunkOffsetsAndInput chunks,
-        AtomicArray<List<InferenceTextEmbeddingByteResults.InferenceByteEmbedding>> debatchedResults
+        AtomicArray<List<InferenceByteEmbedding>> debatchedResults
     ) {
-        var all = new ArrayList<InferenceTextEmbeddingByteResults.InferenceByteEmbedding>();
+        var all = new ArrayList<InferenceByteEmbedding>();
         for (int i = 0; i < debatchedResults.length(); i++) {
             var subBatch = debatchedResults.get(i);
             all.addAll(subBatch);