diff --git a/src/benchmarks/micro/libraries/System.IO.Compression/Zstandard.cs b/src/benchmarks/micro/libraries/System.IO.Compression/Zstandard.cs
new file mode 100644
index 00000000000..2f4ca8132aa
--- /dev/null
+++ b/src/benchmarks/micro/libraries/System.IO.Compression/Zstandard.cs
@@ -0,0 +1,76 @@
+
+// Licensed to the .NET Foundation under one or more agreements.
+// The .NET Foundation licenses this file to you under the MIT license.
+// See the LICENSE file in the project root for more information.
+
+using BenchmarkDotNet.Attributes;
+using MicroBenchmarks;
+
+#if NET11_0_OR_GREATER
+
+namespace System.IO.Compression
+{
+    [BenchmarkCategory(Categories.Libraries, Categories.NoWASM)]
+    public class Zstandard : CompressionStreamPerfTestBase
+    {
+        private const int Window = 22;
+
+        public override Stream CreateStream(Stream stream, CompressionMode mode) => new ZstandardStream(stream, mode);
+        public override Stream CreateStream(Stream stream, CompressionLevel level) => new ZstandardStream(stream, level);
+
+        [Benchmark]
+        public Span<byte> Compress_WithState()
+        {
+            using (ZstandardEncoder encoder = new ZstandardEncoder(GetQuality(level), Window))
+            {
+                Span<byte> output = new Span<byte>(CompressedFile.CompressedData);
+                ReadOnlySpan<byte> input = CompressedFile.UncompressedData;
+                while (!input.IsEmpty && !output.IsEmpty)
+                {
+                    encoder.Compress(input, output, out int bytesConsumed, out int written, isFinalBlock: false);
+                    input = input.Slice(bytesConsumed);
+                    output = output.Slice(written);
+                }
+                encoder.Compress(input, output, out int bytesConsumed2, out int written2, isFinalBlock: true);
+
+                return output;
+            }
+        }
+
+        [Benchmark]
+        public Span<byte> Decompress_WithState() // the level argument is not used here, but it describes how the data was compressed (in the benchmark id)
+        {
+            using (ZstandardDecoder decoder = new ZstandardDecoder())
+            {
+                Span<byte> output = new Span<byte>(CompressedFile.UncompressedData);
+                ReadOnlySpan<byte> input = CompressedFile.CompressedData;
+                while (!input.IsEmpty && !output.IsEmpty)
+                {
+                    decoder.Decompress(input, output, out int bytesConsumed, out int written);
+                    input = input.Slice(bytesConsumed);
+                    output = output.Slice(written);
+                }
+
+                return output;
+            }
+        }
+
+        [Benchmark]
+        [MemoryRandomization]
+        public bool Compress_WithoutState()
+            => ZstandardEncoder.TryCompress(CompressedFile.UncompressedData, CompressedFile.CompressedData, out int bytesWritten, GetQuality(level), Window);
+
+        /// <summary>
+        /// The perf tests for the instant decompression aren't exactly indicative of real-world scenarios since they require you to know 
+        /// either the exact figure or the upper bound of the uncompressed size of your given compressed data.
+        /// </summary>
+        [Benchmark]
+        public bool Decompress_WithoutState() // the level argument is not used here, but it describes how the data was compressed (in the benchmark id)
+            => ZstandardDecoder.TryDecompress(CompressedFile.CompressedData, CompressedFile.UncompressedData, out int bytesWritten);
+
+        private static int GetQuality(CompressionLevel compressLevel)
+            => compressLevel == CompressionLevel.Optimal ? 11 : compressLevel == CompressionLevel.Fastest ? 1 : 0;
+    }
+}
+
+#endif
\ No newline at end of file