diff --git a/js/web/lib/wasm/jsep/webnn/webnn.d.ts b/js/web/lib/wasm/jsep/webnn/webnn.d.ts
index 5cb0f4e74c3df..3505772cd2b73 100644
--- a/js/web/lib/wasm/jsep/webnn/webnn.d.ts
+++ b/js/web/lib/wasm/jsep/webnn/webnn.d.ts
@@ -13,7 +13,6 @@ type MLPowerPreference = 'default'|'high-performance'|'low-power';
 interface MLContextOptions {
   deviceType?: MLDeviceType;
   powerPreference?: MLPowerPreference;
-  numThreads?: number;
 }
 interface ML {
   createContext(options?: MLContextOptions): Promise<MLContext>;
diff --git a/js/web/lib/wasm/wasm-core-impl.ts b/js/web/lib/wasm/wasm-core-impl.ts
index 0668ac1931988..5f219f63aaf61 100644
--- a/js/web/lib/wasm/wasm-core-impl.ts
+++ b/js/web/lib/wasm/wasm-core-impl.ts
@@ -297,14 +297,13 @@ export const createSession = async (
           const context = (webnnOptions as InferenceSession.WebNNOptionsWithMLContext)?.context;
           const gpuDevice = (webnnOptions as InferenceSession.WebNNOptionsWebGpu)?.gpuDevice;
           const deviceType = (webnnOptions as InferenceSession.WebNNContextOptions)?.deviceType;
-          const numThreads = (webnnOptions as InferenceSession.WebNNContextOptions)?.numThreads;
           const powerPreference = (webnnOptions as InferenceSession.WebNNContextOptions)?.powerPreference;
           if (context) {
             wasm.currentContext = context as MLContext;
           } else if (gpuDevice) {
             wasm.currentContext = await navigator.ml.createContext(gpuDevice);
           } else {
-            wasm.currentContext = await navigator.ml.createContext({ deviceType, numThreads, powerPreference });
+            wasm.currentContext = await navigator.ml.createContext({ deviceType, powerPreference });
           }
         } else {
           wasm.currentContext = await navigator.ml.createContext();
diff --git a/js/web/test/test-runner.ts b/js/web/test/test-runner.ts
index a8945222b485a..aa62c8dc22c40 100644
--- a/js/web/test/test-runner.ts
+++ b/js/web/test/test-runner.ts
@@ -291,14 +291,9 @@ export class ModelTestContext {
       if (['ml-tensor', 'ml-location'].includes(modelTest.ioBinding)) {
         const webnnOptions = executionProviderConfig as ort.InferenceSession.WebNNExecutionProviderOption;
         const deviceType = (webnnOptions as ort.InferenceSession.WebNNContextOptions)?.deviceType;
-        const numThreads = (webnnOptions as ort.InferenceSession.WebNNContextOptions)?.numThreads;
         const powerPreference = (webnnOptions as ort.InferenceSession.WebNNContextOptions)?.powerPreference;
 
-        mlContext = await navigator.ml.createContext({
-          deviceType,
-          numThreads,
-          powerPreference,
-        });
+        mlContext = await navigator.ml.createContext({ deviceType, powerPreference });
         (executionProviderConfig as ort.InferenceSession.WebNNExecutionProviderOption).context = mlContext;
         if (!deviceType) {
           (executionProviderConfig as ort.InferenceSession.WebNNContextOptions).deviceType = deviceType;