diff --git a/README.md b/README.md
index 84905a85..388194be 100644
--- a/README.md
+++ b/README.md
@@ -118,8 +118,6 @@ export function HelloTriangle() {
       passEncoder.end();
 
       device.queue.submit([commandEncoder.finish()]);
-
-      context.present();
     };
     helloTriangle();
   }, [ref]);
@@ -164,16 +162,8 @@ ctx.canvas.height = ctx.canvas.clientHeight * PixelRatio.get();
 
 ### Frame Scheduling
 
-In React Native, we want to keep frame presentation as a manual operation as we plan to provide more advanced rendering options that are React Native specific.  
-This means that when you are ready to present a frame, you need to call `present` on the context.
-
-```tsx
-// draw
-// submit to the queue
-device.queue.submit([commandEncoder.finish()]);
-// This method is React Native only
-context.present();
-```
+Frames are presented automatically once you submit command buffers to the device queue, matching the behaviour on the web.  
+The native context still exposes a `present()` method for backwards compatibility, but calling it is optional and typically a no-op.
 
 ### External Textures
 
diff --git a/apps/example/src/ThreeJS/Cube.tsx b/apps/example/src/ThreeJS/Cube.tsx
index d3e9707b..46c4154c 100644
--- a/apps/example/src/ThreeJS/Cube.tsx
+++ b/apps/example/src/ThreeJS/Cube.tsx
@@ -31,7 +31,7 @@ export const Cube = () => {
       mesh.rotation.y = time / 1000;
 
       renderer.render(scene, camera);
-      context.present();
+      //context.present();
     }
     renderer.setAnimationLoop(animate);
     return () => {
diff --git a/apps/example/src/components/useWebGPU.ts b/apps/example/src/components/useWebGPU.ts
index ac8a631a..c57d7974 100644
--- a/apps/example/src/components/useWebGPU.ts
+++ b/apps/example/src/components/useWebGPU.ts
@@ -57,7 +57,7 @@ export const useWebGPU = (scene: Scene) => {
         const render = () => {
           const timestamp = Date.now();
           renderScene(timestamp);
-          context.present();
+          //context.present();
           animationFrameId.current = requestAnimationFrame(render);
         };
 
diff --git a/packages/webgpu/README.md b/packages/webgpu/README.md
index 84905a85..388194be 100644
--- a/packages/webgpu/README.md
+++ b/packages/webgpu/README.md
@@ -118,8 +118,6 @@ export function HelloTriangle() {
       passEncoder.end();
 
       device.queue.submit([commandEncoder.finish()]);
-
-      context.present();
     };
     helloTriangle();
   }, [ref]);
@@ -164,16 +162,8 @@ ctx.canvas.height = ctx.canvas.clientHeight * PixelRatio.get();
 
 ### Frame Scheduling
 
-In React Native, we want to keep frame presentation as a manual operation as we plan to provide more advanced rendering options that are React Native specific.  
-This means that when you are ready to present a frame, you need to call `present` on the context.
-
-```tsx
-// draw
-// submit to the queue
-device.queue.submit([commandEncoder.finish()]);
-// This method is React Native only
-context.present();
-```
+Frames are presented automatically once you submit command buffers to the device queue, matching the behaviour on the web.  
+The native context still exposes a `present()` method for backwards compatibility, but calling it is optional and typically a no-op.
 
 ### External Textures
 
diff --git a/packages/webgpu/cpp/rnwgpu/SurfaceRegistry.h b/packages/webgpu/cpp/rnwgpu/SurfaceRegistry.h
index 266fbd8a..f8bf7ecd 100644
--- a/packages/webgpu/cpp/rnwgpu/SurfaceRegistry.h
+++ b/packages/webgpu/cpp/rnwgpu/SurfaceRegistry.h
@@ -1,9 +1,11 @@
 #pragma once
 
 #include <memory>
+#include <optional>
 #include <shared_mutex>
 #include <unordered_map>
 #include <utility>
+#include <vector>
 
 #ifdef __APPLE__
 #include <dispatch/dispatch.h>
@@ -12,6 +14,16 @@
 
 #include "webgpu/webgpu_cpp.h"
 
+#include "api/Canvas.h"
+
+#ifdef __APPLE__
+namespace dawn::native::metal {
+
+void WaitForCommandsToBeScheduled(WGPUDevice device);
+
+} // namespace dawn::native::metal
+#endif
+
 namespace rnwgpu {
 
 struct NativeInfo {
@@ -32,6 +44,14 @@ class SurfaceInfo {
 
   ~SurfaceInfo() { surface = nullptr; }
 
+  struct PresentRequest {
+    wgpu::Device device;
+    wgpu::Surface surface;
+    int width;
+    int height;
+    std::weak_ptr<Canvas> canvas;
+  };
+
   void reconfigure(int newWidth, int newHeight) {
     std::unique_lock<std::shared_mutex> lock(_mutex);
     config.width = newWidth;
@@ -45,6 +65,7 @@ class SurfaceInfo {
     config.width = width;
     config.height = height;
     config.presentMode = wgpu::PresentMode::Fifo;
+    _needsPresent = false;
     _configure();
   }
 
@@ -55,6 +76,7 @@ class SurfaceInfo {
     } else {
       texture = nullptr;
     }
+    _needsPresent = false;
   }
 
   void *switchToOffscreen() {
@@ -72,6 +94,7 @@ class SurfaceInfo {
       texture = config.device.CreateTexture(&textureDesc);
     }
     surface = nullptr;
+    _needsPresent = false;
     return nativeSurface;
   }
 
@@ -110,6 +133,7 @@ class SurfaceInfo {
       surface.Present();
       texture = nullptr;
     }
+    _needsPresent = false;
   }
 
   void resize(int newWidth, int newHeight) {
@@ -118,24 +142,39 @@ class SurfaceInfo {
     height = newHeight;
   }
 
-  void present() {
+  void setCanvas(std::weak_ptr<Canvas> canvas) {
     std::unique_lock<std::shared_mutex> lock(_mutex);
-    if (surface) {
-      surface.Present();
-    }
+    _canvas = std::move(canvas);
   }
 
   wgpu::Texture getCurrentTexture() {
-    std::shared_lock<std::shared_mutex> lock(_mutex);
+    std::unique_lock<std::shared_mutex> lock(_mutex);
     if (surface) {
       wgpu::SurfaceTexture surfaceTexture;
       surface.GetCurrentTexture(&surfaceTexture);
+      _needsPresent = true;
       return surfaceTexture.texture;
     } else {
+      _needsPresent = false;
       return texture;
     }
   }
 
+  std::optional<PresentRequest> takePendingPresent() {
+    std::unique_lock<std::shared_mutex> lock(_mutex);
+    if (!_needsPresent || !surface) {
+      _needsPresent = false;
+      return std::nullopt;
+    }
+    PresentRequest request{.device = config.device,
+                           .surface = surface,
+                           .width = width,
+                           .height = height,
+                           .canvas = _canvas};
+    _needsPresent = false;
+    return request;
+  }
+
   NativeInfo getNativeInfo() {
     std::shared_lock<std::shared_mutex> lock(_mutex);
     return {.nativeSurface = nativeSurface, .width = width, .height = height};
@@ -191,6 +230,7 @@ class SurfaceInfo {
   }
 
   mutable std::shared_mutex _mutex;
+  std::weak_ptr<Canvas> _canvas;
   void *nativeSurface = nullptr;
   wgpu::Surface surface = nullptr;
   wgpu::Texture texture = nullptr;
@@ -198,6 +238,7 @@ class SurfaceInfo {
   wgpu::SurfaceConfiguration config;
   int width;
   int height;
+  bool _needsPresent = false;
 };
 
 class SurfaceRegistry {
@@ -244,6 +285,35 @@ class SurfaceRegistry {
     return info;
   }
 
+  void presentPendingSurfaces() {
+    std::vector<std::shared_ptr<SurfaceInfo>> infosCopy;
+    {
+      std::shared_lock<std::shared_mutex> lock(_mutex);
+      infosCopy.reserve(_registry.size());
+      for (auto &entry : _registry) {
+        infosCopy.push_back(entry.second);
+      }
+    }
+
+    for (auto &info : infosCopy) {
+      auto request = info->takePendingPresent();
+      if (!request.has_value()) {
+        continue;
+      }
+
+      auto presentRequest = std::move(request.value());
+#ifdef __APPLE__
+      dawn::native::metal::WaitForCommandsToBeScheduled(
+          presentRequest.device.Get());
+#endif
+      if (auto canvas = presentRequest.canvas.lock()) {
+        canvas->setClientWidth(presentRequest.width);
+        canvas->setClientHeight(presentRequest.height);
+      }
+      presentRequest.surface.Present();
+    }
+  }
+
 private:
   SurfaceRegistry() = default;
   mutable std::shared_mutex _mutex;
diff --git a/packages/webgpu/cpp/rnwgpu/api/GPUCanvasContext.cpp b/packages/webgpu/cpp/rnwgpu/api/GPUCanvasContext.cpp
index 15e7f4ac..cc10e0a4 100644
--- a/packages/webgpu/cpp/rnwgpu/api/GPUCanvasContext.cpp
+++ b/packages/webgpu/cpp/rnwgpu/api/GPUCanvasContext.cpp
@@ -52,14 +52,17 @@ std::shared_ptr<GPUTexture> GPUCanvasContext::getCurrentTexture() {
 }
 
 void GPUCanvasContext::present() {
+  auto request = _surfaceInfo->takePendingPresent();
+  if (!request.has_value()) {
+    return;
+  }
 #ifdef __APPLE__
   dawn::native::metal::WaitForCommandsToBeScheduled(
-      _surfaceInfo->getDevice().Get());
+      request->device.Get());
 #endif
-  auto size = _surfaceInfo->getSize();
-  _canvas->setClientWidth(size.width);
-  _canvas->setClientHeight(size.height);
-  _surfaceInfo->present();
+  _canvas->setClientWidth(request->width);
+  _canvas->setClientHeight(request->height);
+  request->surface.Present();
 }
 
 } // namespace rnwgpu
diff --git a/packages/webgpu/cpp/rnwgpu/api/GPUCanvasContext.h b/packages/webgpu/cpp/rnwgpu/api/GPUCanvasContext.h
index 976949f8..5c0da0d7 100644
--- a/packages/webgpu/cpp/rnwgpu/api/GPUCanvasContext.h
+++ b/packages/webgpu/cpp/rnwgpu/api/GPUCanvasContext.h
@@ -30,6 +30,7 @@ class GPUCanvasContext : public m::HybridObject {
     auto &registry = rnwgpu::SurfaceRegistry::getInstance();
     _surfaceInfo =
         registry.getSurfaceInfoOrCreate(contextId, _gpu->get(), width, height);
+    _surfaceInfo->setCanvas(_canvas);
   }
 
 public:
diff --git a/packages/webgpu/cpp/rnwgpu/api/GPUQueue.cpp b/packages/webgpu/cpp/rnwgpu/api/GPUQueue.cpp
index 3a7e9ab7..585c7b02 100644
--- a/packages/webgpu/cpp/rnwgpu/api/GPUQueue.cpp
+++ b/packages/webgpu/cpp/rnwgpu/api/GPUQueue.cpp
@@ -6,6 +6,8 @@
 
 #include "Convertors.h"
 
+#include "../SurfaceRegistry.h"
+
 namespace rnwgpu {
 
 struct BufferSource {
@@ -26,6 +28,7 @@ void GPUQueue::submit(
     return;
   }
   _instance.Submit(bufs_size, bufs.data());
+  SurfaceRegistry::getInstance().presentPendingSurfaces();
 }
 
 void GPUQueue::writeBuffer(std::shared_ptr<GPUBuffer> buffer,
diff --git a/packages/webgpu/src/Canvas.tsx b/packages/webgpu/src/Canvas.tsx
index e311f0b4..6bdca1ee 100644
--- a/packages/webgpu/src/Canvas.tsx
+++ b/packages/webgpu/src/Canvas.tsx
@@ -35,7 +35,11 @@ export interface NativeCanvas {
 }
 
 export type RNCanvasContext = GPUCanvasContext & {
-  present: () => void;
+  /**
+   * @deprecated Presentation happens automatically after queue submission.
+   * This method is kept for backwards compatibility and is a no-op.
+   */
+  present?: () => void;
 };
 
 export interface CanvasRef {