diff --git a/BitFaster.Caching.Benchmarks/BitFaster.Caching.Benchmarks.csproj b/BitFaster.Caching.Benchmarks/BitFaster.Caching.Benchmarks.csproj
index ade8391f..d0367bbc 100644
--- a/BitFaster.Caching.Benchmarks/BitFaster.Caching.Benchmarks.csproj
+++ b/BitFaster.Caching.Benchmarks/BitFaster.Caching.Benchmarks.csproj
@@ -16,6 +16,7 @@
   <ItemGroup>
     <PackageReference Include="BenchmarkDotNet" Version="0.12.1" />
     <PackageReference Include="MathNet.Numerics" Version="4.11.0" />
+    <PackageReference Include="Microsoft.Extensions.Caching.Memory" Version="3.1.5" />
     <PackageReference Include="System.Runtime.Caching" Version="4.7.0" />
   </ItemGroup>
 
diff --git a/BitFaster.Caching.Benchmarks/Lru/LruJustGet.cs b/BitFaster.Caching.Benchmarks/Lru/LruJustGet.cs
index fc86b4c7..9859e248 100644
--- a/BitFaster.Caching.Benchmarks/Lru/LruJustGet.cs
+++ b/BitFaster.Caching.Benchmarks/Lru/LruJustGet.cs
@@ -1,11 +1,12 @@
 using BenchmarkDotNet.Attributes;
 using BitFaster.Caching;
+using BitFaster.Caching.Benchmarks.Lru;
 using BitFaster.Caching.Lru;
+using Microsoft.Extensions.Caching.Memory;
 using System;
 using System.Collections.Concurrent;
 using System.Collections.Generic;
 using System.Linq;
-using System.Runtime.Caching;
 using System.Text;
 using System.Threading.Tasks;
 
@@ -23,60 +24,79 @@ public class LruJustGet
         private static readonly FastConcurrentTLru<int, int> fastConcurrentTLru = new FastConcurrentTLru<int, int>(8, 9, EqualityComparer<int>.Default, TimeSpan.FromMinutes(1));
 
         private static readonly int key = 1;
-        private static MemoryCache memoryCache = MemoryCache.Default;
+        private static System.Runtime.Caching.MemoryCache memoryCache = System.Runtime.Caching.MemoryCache.Default;
+
+        Microsoft.Extensions.Caching.Memory.MemoryCache exMemoryCache 
+            = new Microsoft.Extensions.Caching.Memory.MemoryCache(new MemoryCacheOptionsAccessor());
 
         [GlobalSetup]
         public void GlobalSetup()
         {
-            memoryCache.Set(key.ToString(), "test", new CacheItemPolicy());
+            memoryCache.Set(key.ToString(), "test", new System.Runtime.Caching.CacheItemPolicy());
+            exMemoryCache.Set(key, "test");
         }
 
         [Benchmark(Baseline = true)]
-        public void ConcurrentDictionaryGetOrAdd()
+        public void ConcurrentDictionary()
         {
             Func<int, int> func = x => x;
             dictionary.GetOrAdd(1, func);
         }
 
         [Benchmark()]
-        public void FastConcurrentLruGetOrAdd()
+        public void FastConcurrentLru()
         {
             Func<int, int> func = x => x;
             fastConcurrentLru.GetOrAdd(1, func);
         }
 
         [Benchmark()]
-        public void ConcurrentLruGetOrAdd()
+        public void ConcurrentLru()
         {
             Func<int, int> func = x => x;
             concurrentLru.GetOrAdd(1, func);
         }
 
         [Benchmark()]
-        public void FastConcurrentTLruGetOrAdd()
+        public void FastConcurrentTLru()
         {
             Func<int, int> func = x => x;
             fastConcurrentTLru.GetOrAdd(1, func);
         }
 
         [Benchmark()]
-        public void ConcurrentTLruGetOrAdd()
+        public void ConcurrentTLru()
         {
             Func<int, int> func = x => x;
             concurrentTlru.GetOrAdd(1, func);
         }
 
         [Benchmark()]
-        public void ClassicLruGetOrAdd()
+        public void ClassicLru()
         {
             Func<int, int> func = x => x;
             classicLru.GetOrAdd(1, func);
         }
 
         [Benchmark()]
-        public void MemoryCacheGetStringKey()
+        public void RuntimeMemoryCacheGet()
         {
             memoryCache.Get("1");
         }
+
+        [Benchmark()]
+        public void ExtensionsMemoryCacheGet()
+        {
+            exMemoryCache.Get(1);
+        }
+
+        public class MemoryCacheOptionsAccessor
+            : Microsoft.Extensions.Options.IOptions<MemoryCacheOptions>
+        {
+            private readonly MemoryCacheOptions options = new MemoryCacheOptions();
+
+            public MemoryCacheOptions Value => this.options;
+
+        }
     }
 }
diff --git a/README.md b/README.md
index 82700072..3a21457d 100644
--- a/README.md
+++ b/README.md
@@ -158,16 +158,16 @@ In this test the same items are fetched repeatedly, no items are evicted. Repres
 
 FastConcurrentLru does not allocate and is approximately 10x faster than MemoryCache.
 
-|               Method |      Mean |    Error |   StdDev | Ratio |  Gen 0 | Allocated |
-|--------------------- |----------:|---------:|---------:|------:|-------:|----------:|
-| ConcurrentDictionary |  15.06 ns | 0.286 ns | 0.307 ns |  1.00 |      - |         - |
-|    FastConcurrentLru |  20.70 ns | 0.276 ns | 0.258 ns |  1.37 |      - |         - |
-|        ConcurrentLru |  24.09 ns | 0.270 ns | 0.253 ns |  1.60 |      - |         - |
-|   FastConcurrentTLru |  49.57 ns | 0.619 ns | 0.517 ns |  3.30 |      - |         - |
-|       ConcurrentTLru |  64.82 ns | 2.547 ns | 7.391 ns |  4.50 |      - |         - |
-|           ClassicLru |  76.78 ns | 1.412 ns | 3.039 ns |  5.25 |      - |         - |
-|          MemoryCache | 278.37 ns | 3.887 ns | 3.035 ns | 18.50 | 0.0153 |      32 B |
-
+|                Method |      Mean |    Error |   StdDev | Ratio |  Gen 0 | Allocated |
+|---------------------- |----------:|---------:|---------:|------:|-------:|----------:|
+|  ConcurrentDictionary |  16.88 ns | 0.276 ns | 0.245 ns |  1.00 |      - |         - |
+|     FastConcurrentLru |  23.27 ns | 0.491 ns | 0.565 ns |  1.38 |      - |         - |
+|         ConcurrentLru |  26.77 ns | 0.512 ns | 0.666 ns |  1.60 |      - |         - |
+|    FastConcurrentTLru |  54.35 ns | 0.650 ns | 0.576 ns |  3.22 |      - |         - |
+|        ConcurrentTLru |  60.10 ns | 1.024 ns | 1.501 ns |  3.53 |      - |         - |
+|            ClassicLru |  68.04 ns | 1.400 ns | 2.221 ns |  4.12 |      - |         - |
+|    RuntimeMemoryCache | 280.16 ns | 5.607 ns | 7.486 ns | 16.59 | 0.0153 |      32 B |
+| ExtensionsMemoryCache | 342.72 ns | 3.729 ns | 3.114 ns | 20.29 | 0.0114 |      24 B |
 
 ## Meta-programming using structs for JIT dead code removal and inlining