diff --git a/dotnet/src/Connectors/Connectors.Google.UnitTests/Core/VertexAI/VertexAIEmbeddingEndpointTests.cs b/dotnet/src/Connectors/Connectors.Google.UnitTests/Core/VertexAI/VertexAIEmbeddingEndpointTests.cs
new file mode 100644
index 000000000000..c3a9702810f1
--- /dev/null
+++ b/dotnet/src/Connectors/Connectors.Google.UnitTests/Core/VertexAI/VertexAIEmbeddingEndpointTests.cs
@@ -0,0 +1,105 @@
+// Copyright (c) Microsoft. All rights reserved.
+
+using System;
+using System.Collections.Generic;
+using System.Net.Http;
+using System.Threading.Tasks;
+using Microsoft.SemanticKernel.Connectors.Google;
+using Microsoft.SemanticKernel.Connectors.Google.Core;
+using Xunit;
+
+namespace SemanticKernel.Connectors.Google.UnitTests.Core.VertexAI;
+
+public sealed class VertexAIEmbeddingEndpointTests : IDisposable
+{
+    private readonly HttpMessageHandlerStub _messageHandlerStub;
+    private readonly HttpClient _httpClient;
+
+    public VertexAIEmbeddingEndpointTests()
+    {
+        this._messageHandlerStub = new HttpMessageHandlerStub();
+        this._messageHandlerStub.ResponseToReturn.Content = new StringContent(
+            """
+            {
+              "embedding": {
+                "values": [0.1, 0.2, 0.3]
+              }
+            }
+            """);
+        this._httpClient = new HttpClient(this._messageHandlerStub, false);
+    }
+
+    [Theory]
+    [InlineData("gemini-embedding-2", true)]
+    [InlineData("gemini-embedding-2-preview", true)]
+    [InlineData("gemini-embedding-2-0", true)]
+    [InlineData("GEMINI-EMBEDDING-2", true)]
+    [InlineData("gemini-embedding-001", false)]
+    [InlineData("textembedding-gecko", false)]
+    [InlineData("textembedding-gecko@003", false)]
+    [InlineData("text-embedding-004", false)]
+    [InlineData("custom-model", false)]
+    public void UsesEmbedContentEndpoint_ReturnsExpectedValue(string modelId, bool expected)
+    {
+        Assert.Equal(expected, VertexAIEmbeddingClient.UsesEmbedContentEndpoint(modelId));
+    }
+
+    [Fact]
+    public async Task Constructor_UsesEmbedContentSuffix_ForGeminiEmbedding2Async()
+    {
+        // Arrange
+        var client = this.CreateClient("gemini-embedding-2");
+
+        // Act
+        await client.GenerateEmbeddingsAsync(["hello"]);
+
+        // Assert
+        Assert.NotNull(this._messageHandlerStub.RequestUri);
+        Assert.Contains(":embedContent", this._messageHandlerStub.RequestUri.ToString(), StringComparison.Ordinal);
+        Assert.DoesNotContain(":predict", this._messageHandlerStub.RequestUri.ToString(), StringComparison.Ordinal);
+    }
+
+    [Fact]
+    public async Task Constructor_UsesPredictSuffix_ForLegacyModelsAsync()
+    {
+        // Arrange – predict path expects the legacy response shape
+        this._messageHandlerStub.ResponseToReturn.Content = new StringContent(
+            """
+            {
+              "predictions": [
+                {
+                  "embeddings": {
+                    "values": [0.1, 0.2, 0.3]
+                  }
+                }
+              ]
+            }
+            """);
+        var client = this.CreateClient("gemini-embedding-001");
+
+        // Act
+        await client.GenerateEmbeddingsAsync(["hello"]);
+
+        // Assert
+        Assert.NotNull(this._messageHandlerStub.RequestUri);
+        Assert.Contains(":predict", this._messageHandlerStub.RequestUri.ToString(), StringComparison.Ordinal);
+        Assert.DoesNotContain(":embedContent", this._messageHandlerStub.RequestUri.ToString(), StringComparison.Ordinal);
+    }
+
+    public void Dispose()
+    {
+        this._httpClient.Dispose();
+        this._messageHandlerStub.Dispose();
+    }
+
+    private VertexAIEmbeddingClient CreateClient(string modelId)
+    {
+        return new VertexAIEmbeddingClient(
+            httpClient: this._httpClient,
+            modelId: modelId,
+            bearerTokenProvider: () => ValueTask.FromResult("fake-key"),
+            apiVersion: VertexAIVersion.V1,
+            location: "us-central1",
+            projectId: "fake-project-id");
+    }
+}
diff --git a/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbedContentRequest.cs b/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbedContentRequest.cs
new file mode 100644
index 000000000000..ceae25b8e104
--- /dev/null
+++ b/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbedContentRequest.cs
@@ -0,0 +1,38 @@
+// Copyright (c) Microsoft. All rights reserved.
+
+using System.Text.Json.Serialization;
+
+namespace Microsoft.SemanticKernel.Connectors.Google.Core;
+
+/// <summary>
+/// Request body for Vertex AI <c>:embedContent</c> (Gemini Embedding 2 models).
+/// </summary>
+internal sealed class VertexAIEmbedContentRequest
+{
+    [JsonPropertyName("content")]
+    public GeminiContent Content { get; set; } = null!;
+
+    [JsonPropertyName("outputDimensionality")]
+    [JsonIgnore(Condition = JsonIgnoreCondition.WhenWritingNull)]
+    public int? OutputDimensionality { get; set; }
+
+    [JsonPropertyName("taskType")]
+    [JsonIgnore(Condition = JsonIgnoreCondition.WhenWritingNull)]
+    public string? TaskType { get; set; }
+
+    [JsonPropertyName("title")]
+    [JsonIgnore(Condition = JsonIgnoreCondition.WhenWritingNull)]
+    public string? Title { get; set; }
+
+    public static VertexAIEmbedContentRequest FromText(string text, int? dimensions = null) => new()
+    {
+        Content = new GeminiContent
+        {
+            Parts =
+            [
+                new GeminiPart { Text = text }
+            ]
+        },
+        OutputDimensionality = dimensions
+    };
+}
diff --git a/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbedContentResponse.cs b/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbedContentResponse.cs
new file mode 100644
index 000000000000..24180be971cd
--- /dev/null
+++ b/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbedContentResponse.cs
@@ -0,0 +1,23 @@
+// Copyright (c) Microsoft. All rights reserved.
+
+using System;
+using System.Text.Json.Serialization;
+
+namespace Microsoft.SemanticKernel.Connectors.Google.Core;
+
+/// <summary>
+/// Response body for Vertex AI <c>:embedContent</c> (Gemini Embedding 2 models).
+/// </summary>
+internal sealed class VertexAIEmbedContentResponse
+{
+    [JsonPropertyName("embedding")]
+    [JsonRequired]
+    public ResponseEmbedding Embedding { get; set; } = null!;
+
+    internal sealed class ResponseEmbedding
+    {
+        [JsonPropertyName("values")]
+        [JsonRequired]
+        public ReadOnlyMemory<float> Values { get; set; }
+    }
+}
diff --git a/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbeddingClient.cs b/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbeddingClient.cs
index cb59e0087481..4d21aabde548 100644
--- a/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbeddingClient.cs
+++ b/dotnet/src/Connectors/Connectors.Google/Core/VertexAI/VertexAIEmbeddingClient.cs
@@ -1,4 +1,4 @@
-// Copyright (c) Microsoft. All rights reserved.
+// Copyright (c) Microsoft. All rights reserved.
 
 using System;
 using System.Collections.Generic;
@@ -19,6 +19,7 @@ internal sealed class VertexAIEmbeddingClient : ClientBase
     private readonly string _embeddingModelId;
     private readonly Uri _embeddingEndpoint;
     private readonly int? _dimensions;
+    private readonly bool _usesEmbedContent;
 
     /// <summary>
     /// Represents a client for interacting with the embeddings models by Vertex AI.
@@ -54,10 +55,19 @@ public VertexAIEmbeddingClient(
         string baseUri = GetVertexAIBaseUri(location);
 
         this._embeddingModelId = modelId;
-        this._embeddingEndpoint = new Uri($"{baseUri}/{versionSubLink}/projects/{projectId}/locations/{location}/publishers/google/models/{this._embeddingModelId}:predict");
+        this._usesEmbedContent = UsesEmbedContentEndpoint(modelId);
+        string endpointSuffix = this._usesEmbedContent ? "embedContent" : "predict";
+        this._embeddingEndpoint = new Uri($"{baseUri}/{versionSubLink}/projects/{projectId}/locations/{location}/publishers/google/models/{this._embeddingModelId}:{endpointSuffix}");
         this._dimensions = dimensions;
     }
 
+    /// <summary>
+    /// Newer multimodal Gemini Embedding 2 models use the <c>:embedContent</c> endpoint.
+    /// Legacy text embedding models continue to use <c>:predict</c>.
+    /// </summary>
+    internal static bool UsesEmbedContentEndpoint(string modelId)
+        => modelId.StartsWith("gemini-embedding-2", StringComparison.OrdinalIgnoreCase);
+
     /// <summary>
     /// Generates embeddings for the given data asynchronously.
     /// </summary>
@@ -72,8 +82,13 @@ public async Task<IList<ReadOnlyMemory<float>>> GenerateEmbeddingsAsync(
     {
         Verify.NotNullOrEmpty(data);
 
-        var geminiRequest = this.GetEmbeddingRequest(data, options);
-        using var httpRequestMessage = await this.CreateHttpRequestAsync(geminiRequest, this._embeddingEndpoint).ConfigureAwait(false);
+        if (this._usesEmbedContent)
+        {
+            return await this.GenerateEmbedContentEmbeddingsAsync(data, options, cancellationToken).ConfigureAwait(false);
+        }
+
+        var predictRequest = this.GetEmbeddingRequest(data, options);
+        using var httpRequestMessage = await this.CreateHttpRequestAsync(predictRequest, this._embeddingEndpoint).ConfigureAwait(false);
 
         string body = await this.SendRequestAndGetStringBodyAsync(httpRequestMessage, cancellationToken)
             .ConfigureAwait(false);
@@ -81,6 +96,32 @@ public async Task<IList<ReadOnlyMemory<float>>> GenerateEmbeddingsAsync(
         return DeserializeAndProcessEmbeddingsResponse(body);
     }
 
+    /// <summary>
+    /// The Vertex AI <c>:embedContent</c> API accepts a single content object per request.
+    /// Issue one call per input string and preserve order.
+    /// </summary>
+    private async Task<IList<ReadOnlyMemory<float>>> GenerateEmbedContentEmbeddingsAsync(
+        IList<string> data,
+        EmbeddingGenerationOptions? options,
+        CancellationToken cancellationToken)
+    {
+        var results = new List<ReadOnlyMemory<float>>(data.Count);
+        int? dimensions = options?.Dimensions ?? this._dimensions;
+
+        foreach (string text in data)
+        {
+            var request = VertexAIEmbedContentRequest.FromText(text, dimensions);
+            using var httpRequestMessage = await this.CreateHttpRequestAsync(request, this._embeddingEndpoint).ConfigureAwait(false);
+
+            string body = await this.SendRequestAndGetStringBodyAsync(httpRequestMessage, cancellationToken)
+                .ConfigureAwait(false);
+
+            results.Add(DeserializeAndProcessEmbedContentResponse(body));
+        }
+
+        return results;
+    }
+
     private VertexAIEmbeddingRequest GetEmbeddingRequest(IEnumerable<string> data, EmbeddingGenerationOptions? options = null)
         => VertexAIEmbeddingRequest.FromData(data, options?.Dimensions ?? this._dimensions);
 
@@ -89,4 +130,7 @@ private static List<ReadOnlyMemory<float>> DeserializeAndProcessEmbeddingsRespon
 
     private static List<ReadOnlyMemory<float>> ProcessEmbeddingsResponse(VertexAIEmbeddingResponse embeddingsResponse)
         => embeddingsResponse.Predictions.Select(prediction => prediction.Embeddings.Values).ToList();
+
+    private static ReadOnlyMemory<float> DeserializeAndProcessEmbedContentResponse(string body)
+        => DeserializeResponse<VertexAIEmbedContentResponse>(body).Embedding.Values;
 }