diff --git a/cgmanifests/generated/cgmanifest.json b/cgmanifests/generated/cgmanifest.json
index 044588c080072..46f8c8891dda5 100644
--- a/cgmanifests/generated/cgmanifest.json
+++ b/cgmanifests/generated/cgmanifest.json
@@ -346,7 +346,7 @@
       "component": {
         "type": "git",
         "git": {
-          "commitHash": "12a3b24c456cebd9fd11f23ac0164f78129b00c6",
+          "commitHash": "b9b4a37041dec3dd62ac92014a6cc1aece48d9f3",
           "repositoryUrl": "https://github.com/google/dawn.git"
         },
         "comments": "dawn"
diff --git a/cmake/deps.txt b/cmake/deps.txt
index d1a528bd6b4be..c73d9a4e3532f 100644
--- a/cmake/deps.txt
+++ b/cmake/deps.txt
@@ -58,5 +58,5 @@ extensions;https://github.com/microsoft/onnxruntime-extensions/archive/f3f6caa6e
 composable_kernel;https://github.com/ROCmSoftwarePlatform/composable_kernel/archive/204da9c522cebec5220bba52cd3542ebcaf99e7a.zip;1827348efd47831c13074245274d41b7cae8a557
 directx_headers;https://github.com/microsoft/DirectX-Headers/archive/refs/tags/v1.613.1.zip;47653509a3371eabb156360f42faf582f314bf2e
 cudnn_frontend;https://github.com/NVIDIA/cudnn-frontend/archive/refs/tags/v1.7.0.zip;d0753d8d5b39947ca0729d7773cb84653a129eb1
-dawn;https://github.com/google/dawn/archive/12a3b24c456cebd9fd11f23ac0164f78129b00c6.zip;ad428f6dc16f1336d584f7bad5714e1097dafc43
+dawn;https://github.com/google/dawn/archive/b9b4a37041dec3dd62ac92014a6cc1aece48d9f3.zip;e8b8c2ebabdedb7c57d931fc4a19ae22146d31e1
 kleidiai;https://gitlab.arm.com/kleidi/kleidiai/-/archive/d15722976120710080ca098fe8ddabf4556cb40f/kleidiai-d15722976120710080ca098fe8ddabf4556cb40f.zip;d6c840d00c3b05aedf06e957ddaece1013d1f40b
diff --git a/cmake/external/onnxruntime_external_deps.cmake b/cmake/external/onnxruntime_external_deps.cmake
index 761ce47582ee5..e95656969866f 100644
--- a/cmake/external/onnxruntime_external_deps.cmake
+++ b/cmake/external/onnxruntime_external_deps.cmake
@@ -631,7 +631,7 @@ if (onnxruntime_USE_WEBGPU)
       URL_HASH SHA1=${DEP_SHA1_dawn}
       # All previous patches are merged into the upstream dawn project. We don't need to apply any patches right now.
       # if we need to apply patches in the future, we can uncomment the following line.
-      PATCH_COMMAND ${Patch_EXECUTABLE} --binary --ignore-whitespace -p1 < ${PROJECT_SOURCE_DIR}/patches/dawn/dawn.patch
+      # PATCH_COMMAND ${Patch_EXECUTABLE} --binary --ignore-whitespace -p1 < ${PROJECT_SOURCE_DIR}/patches/dawn/dawn.patch
     )
   endif()
 
diff --git a/cmake/patches/dawn/dawn.patch b/cmake/patches/dawn/dawn.patch
deleted file mode 100644
index 868db0c43e3a5..0000000000000
--- a/cmake/patches/dawn/dawn.patch
+++ /dev/null
@@ -1,118 +0,0 @@
-diff --git a/src/emdawnwebgpu/CMakeLists.txt b/src/emdawnwebgpu/CMakeLists.txt
-index de673537d3..c98dc46de7 100644
---- a/src/emdawnwebgpu/CMakeLists.txt
-+++ b/src/emdawnwebgpu/CMakeLists.txt
-@@ -78,6 +78,7 @@ if (${DAWN_ENABLE_EMSCRIPTEN})
-         endif()
- 
-         set(ARGS
-+            ${Python3_EXECUTABLE}
-             "${DAWN_EMSCRIPTEN_TOOLCHAIN}/tools/maint/gen_struct_info.py"
-             -q
-             "${EM_BUILD_GEN_DIR}/struct_info_webgpu.json"
-diff --git a/third_party/emdawnwebgpu/library_webgpu.js b/third_party/emdawnwebgpu/library_webgpu.js
-index d1835cb090..df03ea2f94 100644
---- a/third_party/emdawnwebgpu/library_webgpu.js
-+++ b/third_party/emdawnwebgpu/library_webgpu.js
-@@ -16,10 +16,19 @@
-     throw new Error("To use Dawn's library_webgpu.js, disable -sUSE_WEBGPU and first include Dawn's library_webgpu_struct_info.js and library_webgpu_enum_tables.js (before library_webgpu.js)");
-   }
- 
-+  if (MEMORY64) {
-+    throw new Error("The current implementation of Dawn's library_webgpu.js does not support MEMORY64 yet");
-+  }
-+
-   // Helper functions for code generation
-   globalThis.gpu = {
--    convertSentinelToUndefined: function(name) {
--      return `if (${name} == -1) ${name} = undefined;`;
-+    convertSentinelToUndefined: function(name, isPtr = false) {
-+      // When `CAN_ADDRESS_2GB` is true, value `-1` is normalized to `0xFFFFFFFF` for pointer.
-+      if (CAN_ADDRESS_2GB && isPtr) {
-+        return `if (${name} == 0xFFFFFFFF) ${name} = undefined;`;
-+      } else {
-+        return `if (${name} == -1) ${name} = undefined;`;
-+      }
-     },
- 
-     makeGetBool: function(struct, offset) {
-@@ -700,6 +709,7 @@ var LibraryWebGPU = {
-     {{{ makeSetValue('info', C_STRUCTS.WGPUAdapterInfo.adapterType, 'adapterType', 'i32') }}};
-     {{{ makeSetValue('info', C_STRUCTS.WGPUAdapterInfo.vendorID, '0', 'i32') }}};
-     {{{ makeSetValue('info', C_STRUCTS.WGPUAdapterInfo.deviceID, '0', 'i32') }}};
-+    return 1;
-   },
- 
-   wgpuAdapterGetLimits: (adapterPtr, limitsOutPtr) => {
-@@ -882,7 +892,7 @@ var LibraryWebGPU = {
- 
-     if (size === 0) warnOnce('getMappedRange size=0 no longer means WGPU_WHOLE_MAP_SIZE');
- 
--    {{{ gpu.convertSentinelToUndefined('size') }}}
-+    {{{ gpu.convertSentinelToUndefined('size', true) }}}
- 
-     var mapped;
-     try {
-@@ -909,7 +919,7 @@ var LibraryWebGPU = {
- 
-     if (size === 0) warnOnce('getMappedRange size=0 no longer means WGPU_WHOLE_MAP_SIZE');
- 
--    {{{ gpu.convertSentinelToUndefined('size') }}}
-+    {{{ gpu.convertSentinelToUndefined('size', true) }}}
- 
-     var mapped;
-     try {
-@@ -950,7 +960,7 @@ var LibraryWebGPU = {
-     var buffer = WebGPU.getJsObject(bufferPtr);
-     WebGPU.Internals.bufferOnUnmaps[bufferPtr] = [];
- 
--    {{{ gpu.convertSentinelToUndefined('size') }}}
-+    {{{ gpu.convertSentinelToUndefined('size', true) }}}
- 
-     {{{ runtimeKeepalivePush() }}}
-     WebGPU.Internals.futureInsert(futureId, buffer.mapAsync(mode, offset, size).then(() => {
-@@ -1145,7 +1155,7 @@ var LibraryWebGPU = {
- 
-   wgpuCommandEncoderClearBuffer: (encoderPtr, bufferPtr, offset, size) => {
-     var commandEncoder = WebGPU.getJsObject(encoderPtr);
--    {{{ gpu.convertSentinelToUndefined('size') }}}
-+    {{{ gpu.convertSentinelToUndefined('size', true) }}}
- 
-     var buffer = WebGPU.getJsObject(bufferPtr);
-     commandEncoder.clearBuffer(buffer, offset, size);
-@@ -2103,7 +2113,7 @@ var LibraryWebGPU = {
-   wgpuRenderBundleEncoderSetIndexBuffer: (passPtr, bufferPtr, format, offset, size) => {
-     var pass = WebGPU.getJsObject(passPtr);
-     var buffer = WebGPU.getJsObject(bufferPtr);
--    {{{ gpu.convertSentinelToUndefined('size') }}}
-+    {{{ gpu.convertSentinelToUndefined('size', true) }}}
-     pass.setIndexBuffer(buffer, WebGPU.IndexFormat[format], offset, size);
-   },
- 
-@@ -2116,7 +2126,7 @@ var LibraryWebGPU = {
-   wgpuRenderBundleEncoderSetVertexBuffer: (passPtr, slot, bufferPtr, offset, size) => {
-     var pass = WebGPU.getJsObject(passPtr);
-     var buffer = WebGPU.getJsObject(bufferPtr);
--    {{{ gpu.convertSentinelToUndefined('size') }}}
-+    {{{ gpu.convertSentinelToUndefined('size', true) }}}
-     pass.setVertexBuffer(slot, buffer, offset, size);
-   },
- 
-@@ -2211,7 +2221,7 @@ var LibraryWebGPU = {
-   wgpuRenderPassEncoderSetIndexBuffer: (passPtr, bufferPtr, format, offset, size) => {
-     var pass = WebGPU.getJsObject(passPtr);
-     var buffer = WebGPU.getJsObject(bufferPtr);
--    {{{ gpu.convertSentinelToUndefined('size') }}}
-+    {{{ gpu.convertSentinelToUndefined('size', true) }}}
-     pass.setIndexBuffer(buffer, WebGPU.IndexFormat[format], offset, size);
-   },
- 
-@@ -2234,7 +2244,7 @@ var LibraryWebGPU = {
-   wgpuRenderPassEncoderSetVertexBuffer: (passPtr, slot, bufferPtr, offset, size) => {
-     var pass = WebGPU.getJsObject(passPtr);
-     var buffer = WebGPU.getJsObject(bufferPtr);
--    {{{ gpu.convertSentinelToUndefined('size') }}}
-+    {{{ gpu.convertSentinelToUndefined('size', true) }}}
-     pass.setVertexBuffer(slot, buffer, offset, size);
-   },
- 
diff --git a/onnxruntime/core/providers/webgpu/program_manager.cc b/onnxruntime/core/providers/webgpu/program_manager.cc
index 109bac34d6503..1fdd312d4f0d8 100644
--- a/onnxruntime/core/providers/webgpu/program_manager.cc
+++ b/onnxruntime/core/providers/webgpu/program_manager.cc
@@ -147,16 +147,16 @@ Status ProgramManager::Build(const ProgramBase& program,
     }
   }
 
-  wgpu::ProgrammableStageDescriptor compute_stage{};
-  compute_stage.module = shader_module;
-  compute_stage.entryPoint = "main";
+  wgpu::ComputeState compute_state{};
+  compute_state.module = shader_module;
+  compute_state.entryPoint = "main";
   if (!constant_entries.empty()) {
-    compute_stage.constants = constant_entries.data();
-    compute_stage.constantCount = constant_entries.size();
+    compute_state.constants = constant_entries.data();
+    compute_state.constantCount = constant_entries.size();
   }
 
   wgpu::ComputePipelineDescriptor pipeline_descriptor{};
-  pipeline_descriptor.compute = compute_stage;
+  pipeline_descriptor.compute = compute_state;
 #ifndef NDEBUG  // if debug build
   pipeline_descriptor.label = program.Name().c_str();
 #endif
diff --git a/onnxruntime/core/providers/webgpu/webgpu_context.cc b/onnxruntime/core/providers/webgpu/webgpu_context.cc
index f7d94207019da..99a645878cd7e 100644
--- a/onnxruntime/core/providers/webgpu/webgpu_context.cc
+++ b/onnxruntime/core/providers/webgpu/webgpu_context.cc
@@ -115,12 +115,12 @@ void WebGpuContext::Initialize(const WebGpuBufferCacheConfig& buffer_cache_confi
       device_desc.requiredLimits = &required_limits;
 
       // TODO: revise temporary error handling
-      device_desc.SetUncapturedErrorCallback([](const wgpu::Device& /*device*/, wgpu::ErrorType type, const char* message) {
-        LOGS_DEFAULT(ERROR) << "WebGPU device error(" << int(type) << "): " << message;
+      device_desc.SetUncapturedErrorCallback([](const wgpu::Device& /*device*/, wgpu::ErrorType type, wgpu::StringView message) {
+        LOGS_DEFAULT(ERROR) << "WebGPU device error(" << int(type) << "): " << std::string_view{message};
       });
       // TODO: revise temporary device lost handling
-      device_desc.SetDeviceLostCallback(wgpu::CallbackMode::AllowSpontaneous, [](const wgpu::Device& /*device*/, wgpu::DeviceLostReason reason, const char* message) {
-        LOGS_DEFAULT(INFO) << "WebGPU device lost (" << int(reason) << "): " << message;
+      device_desc.SetDeviceLostCallback(wgpu::CallbackMode::AllowSpontaneous, [](const wgpu::Device& /*device*/, wgpu::DeviceLostReason reason, wgpu::StringView message) {
+        LOGS_DEFAULT(INFO) << "WebGPU device lost (" << int(reason) << "): " << std::string_view{message};
       });
 
       ORT_ENFORCE(wgpu::WaitStatus::Success == instance_.WaitAny(adapter_.RequestDevice(
diff --git a/tools/ci_build/github/azure-pipelines/templates/download-deps.yml b/tools/ci_build/github/azure-pipelines/templates/download-deps.yml
index 1a53ce6a423b6..fe3bc60c83dea 100644
--- a/tools/ci_build/github/azure-pipelines/templates/download-deps.yml
+++ b/tools/ci_build/github/azure-pipelines/templates/download-deps.yml
@@ -11,7 +11,7 @@ steps:
       packageType: upack
       feed: '/7424c8e4-5c62-490e-95c4-79446f31017c'
       definition: '517c4f6f-5437-4392-a70d-4f15ec5be2f0'
-      version: 1.0.208
+      version: 1.0.213
       downloadPath: $(Build.BinariesDirectory)/deps
 
 # The private ADO project
@@ -22,7 +22,7 @@ steps:
       packageType: upack
       feed: '/4c7631f5-24c0-4307-8822-1aa8f180c325'
       definition: 'fd9dd5ad-b73e-4678-890e-edcf680dbc1a'
-      version: 1.0.208
+      version: 1.0.213
       downloadPath: $(Build.BinariesDirectory)/deps
 
 # You can add more ADO accounts at here.