![François Beaufort](https://web.dev/images/authors/beaufortfrancois.jpg) François Beaufort [GitHub](https://github.com/beaufortfrancois)

<br />

## Support for 16-bit floating-point values in WGSL

In WGSL, the [`f16` type](https://www.w3.org/TR/WGSL/#f16) is the set of 16-bit floating-point values of the IEEE-754 binary16 (half precision) format. It means that it uses 16 bits to represent a floating-point number, as opposed to 32 bits for conventional single-precision floating-point ([`f32`](https://gpuweb.github.io/gpuweb/wgsl/#f32)). This smaller size can lead to significant [performance improvements](https://therealmjp.github.io/posts/shader-fp16/), especially when processing large amounts of data.

For comparison, on an Apple M1 Pro device, the `f16` implementation of [Llama2 7B models](https://huggingface.co/meta-llama/Llama-2-7b) used in the [WebLLM chat demo](https://webllm.mlc.ai/#chat-demo) is significantly faster than the `f32` implementation, with a 28% improvement in prefill speed and a 41% improvement in decoding speed as shown in the following screenshots.
![Screenshot of WebLLM chat demos with f32 and f16 Llama2 7B models.](https://developer.chrome.com/static/blog/new-in-webgpu-120/image/webllm-chat-demos.png) WebLLM chat demos with `f32` (left) and `f16` (right) Llama2 7B models.

Not all GPUs support 16-bit floating-point values. When the [`"shader-f16"`](https://gpuweb.github.io/gpuweb/#shader-f16) feature is available in a `GPUAdapter`, you can now request a `GPUDevice` with this feature and create a WGSL shader module that takes advantage of the half-precision floating-point type `f16`. This type is valid to use in the WGSL shader module only if you enable the [`f16` WGSL extension](https://gpuweb.github.io/gpuweb/wgsl/#extension-f16) with `enable f16;`. Otherwise, [createShaderModule()](https://developer.mozilla.org/en-US/docs/Web/API/GPUDevice/createShaderModule) will generate a validation error. See the following minimal example and [issue dawn:1510](https://bugs.chromium.org/p/dawn/issues/detail?id=1510).

```javascript
const adapter = await navigator.gpu.requestAdapter();
if (!adapter.features.has("shader-f16")) {
  throw new Error("16-bit floating-point value support is not available");
}
// Explicitly request 16-bit floating-point value support.
const device = await adapter.requestDevice({
  requiredFeatures: ["shader-f16"],
});

const code = `
  enable f16;

  @compute @workgroup_size(1)
  fn main() {
    const c : vec3h = vec3<f16>(1.0h, 2.0h, 3.0h);
  }
`;

const shaderModule = device.createShaderModule({ code });
// Create a compute pipeline with this shader module
// and run the shader on the GPU...
```

It's possible to support both `f16` and `f32` types in the WGSL shader module code with an [`alias`](https://gpuweb.github.io/gpuweb/wgsl/#type-aliases) depending on the `"shader-f16"` feature support as shown in the following snippet.

```javascript
const adapter = await navigator.gpu.requestAdapter();
const hasShaderF16 = adapter.features.has("shader-f16");

const device = await adapter.requestDevice({
  requiredFeatures: hasShaderF16 ? ["shader-f16"] : [],
});

const header = hasShaderF16
  ? `enable f16;
     alias min16float = f16;`
  : `alias min16float = f32;`;

const code = `
  ${header}

  @compute @workgroup_size(1)
  fn main() {
    const c = vec3<min16float>(1.0, 2.0, 3.0);
  }
`;
```

## Push the limits

The maximum number of bytes necessary to hold one sample (pixel or subpixel) of render pipeline output data, across all color attachments, is 32 bytes by default. It is now possible to request up to 64 by using the [`maxColorAttachmentBytesPerSample`](https://gpuweb.github.io/gpuweb/#dom-supported-limits-maxcolorattachmentbytespersample) limit. See the following example and [issue dawn:2036](https://bugs.chromium.org/p/dawn/issues/detail?id=2036).

```javascript
const adapter = await navigator.gpu.requestAdapter();

if (adapter.limits.maxColorAttachmentBytesPerSample < 64) {
  // When the desired limit isn't supported, take action to either fall back to
  // a code path that does not require the higher limit or notify the user that
  // their device does not meet minimum requirements.
}

// Request highest limit of max color attachments bytes per sample.
const device = await adapter.requestDevice({
  requiredLimits: { maxColorAttachmentBytesPerSample: 64 },
});
```

The [`maxInterStageShaderVariables`](https://gpuweb.github.io/gpuweb/#dom-supported-limits-maxinterstageshadervariables) and [`maxInterStageShaderComponents`](https://gpuweb.github.io/gpuweb/#dom-supported-limits-maxinterstageshadercomponents) limits used for inter-stage communication have been increased on all platforms. See [issue dawn:1448](https://bugs.chromium.org/p/dawn/issues/detail?id=1448) for details.

For each shader stage, the maximum number of bind group layout entries across a pipeline layout which are storage buffers is 8 by default. It is now possible to request up to 10 by using the [`maxStorageBuffersPerShaderStage`](https://gpuweb.github.io/gpuweb/#dom-supported-limits-maxstoragebufferspershaderstage) limit. See [issue dawn:2159](https://bugs.chromium.org/p/dawn/issues/detail?id=2159).

A new [`maxBindGroupsPlusVertexBuffers`](https://gpuweb.github.io/gpuweb/#dom-supported-limits-maxbindgroupsplusvertexbuffers) limit has been added. It consists of the maximum number of bind group and vertex buffer slots used simultaneously, counting any empty slots below the highest index. Its default value is 24. See [issue dawn:1849](https://bugs.chromium.org/p/dawn/issues/detail?id=1849).

## Changes to depth-stencil state

To improve the developer experience, the depth-stencil state [`depthWriteEnabled`](https://gpuweb.github.io/gpuweb/#dom-gpudepthstencilstate-depthwriteenabled) and [`depthCompare`](https://gpuweb.github.io/gpuweb/#dom-gpudepthstencilstate-depthcompare) attributes are not always required anymore: `depthWriteEnabled` is required only for formats with depth, and `depthCompare` is not required for formats with depth if not used at all. See [issue dawn:2132](https://bugs.chromium.org/p/dawn/issues/detail?id=2132).

## Adapter information updates

Non-standard `type` and `backend` adapter info attributes are now available upon calling [requestAdapterInfo()](https://developer.mozilla.org/en-US/docs/Web/API/GPUAdapter/requestAdapterInfo) when the user has enabled the "WebGPU Developer Features" [flag](https://developer.chrome.com/docs/web-platform/chrome-flags.md.txt#chromeflags) at `chrome://flags/#enable-webgpu-developer-features`. The `type` can be "discrete GPU", "integrated GPU", "CPU", or "unknown". The `backend` is either "WebGPU", "D3D11", "D3D12", "metal", "vulkan", "openGL", "openGLES", or "null". See [issue dawn:2112](https://bugs.chromium.org/p/dawn/issues/detail?id=2112) and [issue dawn:2107](https://bugs.chromium.org/p/dawn/issues/detail?id=2107).
![Screenshot of https://webgpureport.org featuring backend and type in adapter info.](https://developer.chrome.com/static/blog/new-in-webgpu-120/image/webgpureport.png) Adapter info backend and type shown on [https://webgpureport.org](https://webgpureport.org/).

The optional `unmaskHints` list parameter in [requestAdapterInfo()](https://developer.mozilla.org/en-US/docs/Web/API/GPUAdapter/requestAdapterInfo) has been removed. See [issue dawn:1427](https://bugs.chromium.org/p/dawn/issues/detail?id=1427).

## Timestamp queries quantization

Timestamp queries allow applications to measure the execution time of GPU commands with nanosecond precision. However, the WebGPU specification makes timestamp queries optional due to [timing attack](https://gpuweb.github.io/gpuweb/#security-timing) concerns. The Chrome team believes that quantizing timestamp queries provides a good compromise between precision and security, by reducing the resolution to 100 microseconds. See [issue dawn:1800](https://bugs.chromium.org/p/dawn/issues/detail?id=1800).

In Chrome, users can disable timestamp quantization by enabling the "WebGPU Developer Features" [flag](https://developer.chrome.com/docs/web-platform/chrome-flags.md.txt#chromeflags) at `chrome://flags/#enable-webgpu-developer-features`. Note that this flag alone does not enable the [`"timestamp-query"`](https://gpuweb.github.io/gpuweb/#timestamp-query) feature. Its implementation is still experimental and therefore requires the "Unsafe WebGPU Support" flag at `chrome://flags/#enable-unsafe-webgpu`.

In Dawn, a new [device toggle](https://dawn.googlesource.com/dawn.git/+/refs/heads/main/docs/dawn/device_facilities.md#toggles) called "timestamp_quantization" has been added and is enabled by default. The following snippet shows you how to allow the experimental "timestamp-query" feature with no timestamp quantization when requesting a device.

```c++
wgpu::DawnTogglesDescriptor deviceTogglesDesc = {};

const char* allowUnsafeApisToggle = "allow_unsafe_apis";
deviceTogglesDesc.enabledToggles = &allowUnsafeApisToggle;
deviceTogglesDesc.enabledToggleCount = 1;

const char* timestampQuantizationToggle = "timestamp_quantization";
deviceTogglesDesc.disabledToggles = &timestampQuantizationToggle;
deviceTogglesDesc.disabledToggleCount = 1;

wgpu::DeviceDescriptor desc = {.nextInChain = &deviceTogglesDesc};

// Request a device with no timestamp quantization.
myAdapter.RequestDevice(&desc, myCallback, myUserData);
```

## Spring-cleaning features

The experimental "timestamp-query-inside-passes" feature has been renamed to "chromium-experimental-timestamp-query-inside-passes" to make it clear to developers that this feature is experimental and available only in Chromium-based browsers for now. See [issue dawn:1193](https://bugs.chromium.org/p/dawn/issues/detail?id=1193).

The experimental "pipeline-statistics-query" feature, which was only partially implemented, has been removed because it is no longer being developed. See [issue chromium:1177506](https://bugs.chromium.org/p/chromium/issues/detail?id=1177506).

This covers only some of the key highlights. Check out the exhaustive [list of commits](https://dawn.googlesource.com/dawn/+log/chromium/6045..chromium/6099).

## What's New in WebGPU

A list of everything that has been covered in the [What's New in WebGPU](https://developer.chrome.com/docs/web-platform/webgpu/news.md.txt) series.

### Chrome 155-156

- [WGSL fragment_depth extension](https://developer.chrome.com/blog/new-in-webgpu-155-156.md.txt#wgsl-fragment_depth)
- [Texture compression unaligned](https://developer.chrome.com/blog/new-in-webgpu-155-156.md.txt#texture-compression)
- [srgb-linear and display-p3-linear color spaces](https://developer.chrome.com/blog/new-in-webgpu-155-156.md.txt#srgb-linear-and)
- [Additional snorm10-10-10-2 vertex format](https://developer.chrome.com/blog/new-in-webgpu-155-156.md.txt#additional-snorm10-10-10-2)
- [f16 support on Linux for NVIDIA](https://developer.chrome.com/blog/new-in-webgpu-155-156.md.txt#f16-support)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-155-156.md.txt#dawn-updates)

### Chrome 153-154

- [WGSL buffer_view extension](https://developer.chrome.com/blog/new-in-webgpu-153-154.md.txt#wgsl-buffer_view)
- [WGSL swizzle_assignment extension](https://developer.chrome.com/blog/new-in-webgpu-153-154.md.txt#wgsl-swizzle_assignment)
- [WGSL extension updates](https://developer.chrome.com/blog/new-in-webgpu-153-154.md.txt#wgsl-extension)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-153-154.md.txt#dawn-updates)

### Chrome 151-152

- [Subgroup size control](https://developer.chrome.com/blog/new-in-webgpu-151-152.md.txt#subgroup-size)
- [OperationError for setImmediates validation failures](https://developer.chrome.com/blog/new-in-webgpu-151-152.md.txt#operationerror-setimmediates)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-151-152.md.txt#dawn-updates)

### Chrome 149-150

- [Immediates](https://developer.chrome.com/blog/new-in-webgpu-149-150.md.txt#immediates)
- [Stricter validation for transient attachments](https://developer.chrome.com/blog/new-in-webgpu-149-150.md.txt#stricter_validation_for_transient_attachments)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-149-150.md.txt#dawn_updates)

### Chrome 147-148

- [WGSL linear_indexing extension](https://developer.chrome.com/blog/new-in-webgpu-147-148.md.txt#wgsl_linear_indexing_extension)
- [WebGPU on Linux NVIDIA](https://developer.chrome.com/blog/new-in-webgpu-147-148.md.txt#webgpu_on_linux_nvidia)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-147-148.md.txt#dawn_updates)

### Chrome 146

- [Support WebGPU compatibility mode on OpenGL ES 3.1](https://developer.chrome.com/blog/new-in-webgpu-146.md.txt#support_webgpu_compatibility_mode_on_opengl_es_31)
- [Transient attachments](https://developer.chrome.com/blog/new-in-webgpu-146.md.txt#transient_attachments)
- [WGSL texture_and_sampler_let extension](https://developer.chrome.com/blog/new-in-webgpu-146.md.txt#wgsl_texture_and_sampler_let_extension)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-146.md.txt#dawn_updates)

### Chrome 145

- [WGSL subgroup_uniformity extension](https://developer.chrome.com/blog/new-in-webgpu-145.md.txt#wgsl_subgroup_uniformity_extension)
- [Experimental synchronous buffer mapping in workers](https://developer.chrome.com/blog/new-in-webgpu-145.md.txt#experimental_synchronous_buffer_mapping_in_workers)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-145.md.txt#dawn_updates)

### Chrome 144

- [WGSL subgroup_id extension](https://developer.chrome.com/blog/new-in-webgpu-144.md.txt#wgsl_subgroup_id_extension)
- [WGSL uniform_buffer_standard_layout extension](https://developer.chrome.com/blog/new-in-webgpu-144.md.txt#wgsl_uniform_buffer_standard_layout_extension)
- [WebGPU on Linux](https://developer.chrome.com/blog/new-in-webgpu-144.md.txt#webgpu_on_linux)
- [Faster writeBuffer and writeTexture](https://developer.chrome.com/blog/new-in-webgpu144#faster_writebuffer_and_writetexture)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-144.md.txt#dawn_updates)

### Chrome 143

- [Texture component swizzle](https://developer.chrome.com/blog/new-in-webgpu-143.md.txt#texture_component_swizzle)
- [Remove bgra8unorm read-only storage texture usage](https://developer.chrome.com/blog/new-in-webgpu-143.md.txt#remove_bgra8unorm_read-only_storage_texture_usage)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-143.md.txt#dawn_updates)

### Chrome 142

- [Texture format support capabilities extended](https://developer.chrome.com/blog/new-in-webgpu-142.md.txt#texture_format_support_capabilities_extended)
- [Primitive index in WGSL](https://developer.chrome.com/blog/new-in-webgpu-142.md.txt#primitive_index_in_wgsl)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-142.md.txt#dawn_updates)

### Chrome 141

- [Tint IR completed](https://developer.chrome.com/blog/new-in-webgpu-141.md.txt#tint_ir_completed)
- [Integer range analysis in WGSL compiler](https://developer.chrome.com/blog/new-in-webgpu-141.md.txt#integer_range_analysis_in_wgsl_compiler)
- [SPIR-V 1.4 update for Vulkan backend](https://developer.chrome.com/blog/new-in-webgpu-141.md.txt#spir-v_14_update_for_vulkan_backend)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-141.md.txt#dawn_updates)

### Chrome 140

- [Device requests consume adapter](https://developer.chrome.com/blog/new-in-webgpu-140.md.txt#device_requests_consume_adapter)
- [Shorthand for using texture where texture view is used](https://developer.chrome.com/blog/new-in-webgpu-140.md.txt#shorthand_for_using_texture_where_texture_view_is_used)
- [WGSL textureSampleLevel supports 1D textures](https://developer.chrome.com/blog/new-in-webgpu-140.md.txt#wgsl_texturesamplelevel_supports_1d_textures)
- [Deprecate bgra8unorm read-only storage texture usage](https://developer.chrome.com/blog/new-in-webgpu-140.md.txt#deprecate_bgra8unorm_read-only_storage_texture_usage)
- [Remove GPUAdapter isFallbackAdapter attribute](https://developer.chrome.com/blog/new-in-webgpu-140.md.txt#remove_gpuadapter_isfallbackadapter_attribute)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-140.md.txt#dawn_updates)

### Chrome 139

- [3D texture support for BC and ASTC compressed formats](https://developer.chrome.com/blog/new-in-webgpu-139.md.txt#3d_texture_support_for_bc_and_astc_compressed_formats)
- [New "core-features-and-limits" feature](https://developer.chrome.com/blog/new-in-webgpu-139.md.txt#new_core-features-and-limits_feature)
- [Origin trial for WebGPU compatibility mode](https://developer.chrome.com/blog/new-in-webgpu-139.md.txt#origin_trial_for_webgpu_compatibility_mode)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-139.md.txt#dawn_updates)

### Chrome 138

- [Shorthand for using buffer as a binding resource](https://developer.chrome.com/blog/new-in-webgpu-138.md.txt#shorthand_for_using_buffer_as_a_binding_resource)
- [Size requirement changes for buffers mapped at creation](https://developer.chrome.com/blog/new-in-webgpu-138.md.txt#size_requirement_changes_for_buffers_mapped_at_creation)
- [Architecture report for recent GPUs](https://developer.chrome.com/blog/new-in-webgpu-138.md.txt#architecture_report_for_recent_gpus)
- [Deprecate GPUAdapter isFallbackAdapter attribute](https://developer.chrome.com/blog/new-in-webgpu-138.md.txt#deprecate_gpuadapter_isfallbackadapter_attribute)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-138.md.txt#dawn_updates)

### Chrome 137

- [Use texture view for externalTexture binding](https://developer.chrome.com/blog/new-in-webgpu-137.md.txt#use_texture_view_for_externaltexture_binding)
- [Buffers copy without specifying offsets and size](https://developer.chrome.com/blog/new-in-webgpu-137.md.txt#buffers_copy_without_specifying_offsets_and_size)
- [WGSL workgroupUniformLoad using pointer to atomic](https://developer.chrome.com/blog/new-in-webgpu-137.md.txt#wgsl_workgroupuniformload_using_pointer_to_atomic)
- [GPUAdapterInfo powerPreference attribute](https://developer.chrome.com/blog/new-in-webgpu-137.md.txt#gpuadapterinfo_powerpreference_attribute)
- [Remove GPURequestAdapterOptions compatibilityMode attribute](https://developer.chrome.com/blog/new-in-webgpu-137.md.txt#remove_gpurequestadapteroptions_compatibilitymode_attribute)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-137.md.txt#dawn_updates)

### Chrome 136

- [GPUAdapterInfo isFallbackAdapter attribute](https://developer.chrome.com/blog/new-in-webgpu-136.md.txt#gpuadapterinfo_isfallbackadapter_attribute)
- [Shader compilation time improvements on D3D12](https://developer.chrome.com/blog/new-in-webgpu-136.md.txt#shader_compilation_time_improvements_on_d3d12)
- [Save and copy canvas images](https://developer.chrome.com/blog/new-in-webgpu-136.md.txt#save_and_copy_canvas_images)
- [Lift compatibility mode restrictions](https://developer.chrome.com/blog/new-in-webgpu-136.md.txt#lift_compatibility_mode_restrictions)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-136.md.txt#dawn_updates)

### Chrome 135

- [Allow creating pipeline layout with null bind group layout](https://developer.chrome.com/blog/new-in-webgpu-135.md.txt#allow_creating_pipeline_layout_with_null_bind_group_layout)
- [Allow viewports to extend past the render targets bounds](https://developer.chrome.com/blog/new-in-webgpu-135.md.txt#allow_viewports_to_extend_past_the_render_targets_bounds)
- [Easier access to the experimental compatibility mode on Android](https://developer.chrome.com/blog/new-in-webgpu-135.md.txt#easier_access_to_the_experimental_compatibility_mode_on_android)
- [Remove maxInterStageShaderComponents limit](https://developer.chrome.com/blog/new-in-webgpu-135.md.txt#remove_maxinterstageshadercomponents_limit)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-135.md.txt#dawn_updates)

### Chrome 134

- [Improve machine-learning workloads with subgroups](https://developer.chrome.com/blog/new-in-webgpu-134.md.txt#improve_machine-learning_workloads_with_subgroups)
- [Remove float filterable texture types support as blendable](https://developer.chrome.com/blog/new-in-webgpu-134.md.txt#remove_float_filterable_texture_types_support_as_blendable)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-134.md.txt#dawn_updates)

### Chrome 133

- [Additional unorm8x4-bgra and 1-component vertex formats](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#additional_unorm8x4-bgra_and_1-component_vertex_formats)
- [Allow unknown limits to be requested with undefined value](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#allow_unknown_limits_to_be_requested_with_undefined_value)
- [WGSL alignment rules changes](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#wgsl_alignment_rules_changes)
- [WGSL performance gains with discard](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#wgsl_performance_gains_with_discard)
- [Use VideoFrame displaySize for external textures](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#use_videoframe_displaysize_for_external_textures)
- [Handle images with non-default orientations using copyExternalImageToTexture](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#handle_images_with_non-default_orientations_using_copyexternalimagetotexture)
- [Improving developer experience](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#improving_developer_experience)
- [Enable compatibility mode with featureLevel](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#enable_compatibility_mode_with_featurelevel)
- [Experimental subgroup features cleanup](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#experimental_subgroup_features_cleanup)
- [Deprecate maxInterStageShaderComponents limit](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#deprecate_maxinterstageshadercomponents_limit)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-133.md.txt#dawn_updates)

### Chrome 132

- [Texture view usage](https://developer.chrome.com/blog/new-in-webgpu-132.md.txt#texture_view_usage)
- [32-bit float textures blending](https://developer.chrome.com/blog/new-in-webgpu-132.md.txt#32-bit_float_textures_blending)
- [GPUDevice adapterInfo attribute](https://developer.chrome.com/blog/new-in-webgpu-132.md.txt#gpudevice_adapterinfo_attribute)
- [Configuring canvas context with invalid format throw JavaScript error](https://developer.chrome.com/blog/new-in-webgpu-132.md.txt#configuring_canvas_context_with_invalid_format_throw_javascript_error)
- [Filtering sampler restrictions on textures](https://developer.chrome.com/blog/new-in-webgpu-132.md.txt#filtering_sampler_restrictions_on_textures)
- [Extended subgroups experimentation](https://developer.chrome.com/blog/new-in-webgpu-132.md.txt#extended_subgroups_experimentation)
- [Improving developer experience](https://developer.chrome.com/blog/new-in-webgpu-132.md.txt#improving_developer_experience)
- [Experimental support for 16-bit normalized texture formats](https://developer.chrome.com/blog/new-in-webgpu-132.md.txt#experimental_support_for_16-bit_normalized_texture_formats)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-132.md.txt#dawn_updates)

### Chrome 131

- [Clip distances in WGSL](https://developer.chrome.com/blog/new-in-webgpu-131.md.txt#clip_distances_in_wgsl)
- [GPUCanvasContext getConfiguration()](https://developer.chrome.com/blog/new-in-webgpu-131.md.txt#gpucanvascontext_getconfiguration)
- [Point and line primitives must not have depth bias](https://developer.chrome.com/blog/new-in-webgpu-131.md.txt#point_and_line_primitives_must_not_have_depth_bias)
- [Inclusive scan built-in functions for subgroups](https://developer.chrome.com/blog/new-in-webgpu-131.md.txt#inclusive_scan_built-in_functions_for_subgroups)
- [Experimental support for multi-draw indirect](https://developer.chrome.com/blog/new-in-webgpu-131.md.txt#experimental_support_for_multi-draw_indirect)
- [Shader module compilation option strict math](https://developer.chrome.com/blog/new-in-webgpu-131.md.txt#shader_module_compilation_option_strict_math)
- [Remove GPUAdapter requestAdapterInfo()](https://developer.chrome.com/blog/new-in-webgpu-131.md.txt#remove_gpuadapter_requestadapterinfo)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-131.md.txt#dawn_updates)

### Chrome 130

- [Dual source blending](https://developer.chrome.com/blog/new-in-webgpu-130.md.txt#dual_source_blending)
- [Shader compilation time improvements on Metal](https://developer.chrome.com/blog/new-in-webgpu-130.md.txt#shader_compilation_time_improvements_on_metal)
- [Deprecation of GPUAdapter requestAdapterInfo()](https://developer.chrome.com/blog/new-in-webgpu-130.md.txt#deprecation_of_gpuadapter_requestadapterinfo)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-130.md.txt#dawn_updates)

### Chrome 129

- [HDR support with canvas tone mapping mode](https://developer.chrome.com/blog/new-in-webgpu-129.md.txt#hdr_support_with_canvas_tone_mapping_mode)
- [Expanded subgroups support](https://developer.chrome.com/blog/new-in-webgpu-129.md.txt#expanded_subgroups_support)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-129.md.txt#dawn_updates)

### Chrome 128

- [Experimenting with subgroups](https://developer.chrome.com/blog/new-in-webgpu-128.md.txt#experimenting_with_subgroups)
- [Deprecate setting depth bias for lines and points](https://developer.chrome.com/blog/new-in-webgpu-128.md.txt#deprecate_setting_depth_bias_for_lines_and_points)
- [Hide uncaptured error DevTools warning if preventDefault](https://developer.chrome.com/blog/new-in-webgpu-128.md.txt#hide_uncaptured_error_devtools_warning_if_preventdefault)
- [WGSL interpolate sampling first and either](https://developer.chrome.com/blog/new-in-webgpu-128.md.txt#wgsl_interpolate_sampling_first_and_either)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-128.md.txt#dawn_updates)

### Chrome 127

- [Experimental support for OpenGL ES on Android](https://developer.chrome.com/blog/new-in-webgpu-127.md.txt#experimental_support_for_opengl_es_on_android)
- [GPUAdapter info attribute](https://developer.chrome.com/blog/new-in-webgpu-127.md.txt#gpuadapter_info_attribute)
- [WebAssembly interop improvements](https://developer.chrome.com/blog/new-in-webgpu-127.md.txt#webassembly_interop_improvements)
- [Improved command encoder errors](https://developer.chrome.com/blog/new-in-webgpu-127.md.txt#improved_command_encoder_errors)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-127.md.txt#dawn_updates)

### Chrome 126

- [Increase maxTextureArrayLayers limit](https://developer.chrome.com/blog/new-in-webgpu-126.md.txt#increase_maxtexturearraylayers_limit)
- [Buffer upload optimization for Vulkan backend](https://developer.chrome.com/blog/new-in-webgpu-126.md.txt#buffer_upload_optimization_for_vulkan_backend)
- [Shader compilation time improvements](https://developer.chrome.com/blog/new-in-webgpu-126.md.txt#shader_compilation_time_improvements)
- [Submitted command buffers must be unique](https://developer.chrome.com/blog/new-in-webgpu-126.md.txt#submitted_command_buffers_must_be_unique)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-126.md.txt#dawn_updates)

### Chrome 125

- [Subgroups (feature in development)](https://developer.chrome.com/blog/new-in-webgpu-125.md.txt#subgroups_feature_in_development)
- [Render to slice of 3D texture](https://developer.chrome.com/blog/new-in-webgpu-125.md.txt#render_to_slice_of_3d_texture)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-125.md.txt#dawn_updates)

### Chrome 124

- [Read-only and read-write storage textures](https://developer.chrome.com/blog/new-in-webgpu-124.md.txt#read-only_and_read-write_storage_textures)
- [Service workers and shared workers support](https://developer.chrome.com/blog/new-in-webgpu-124.md.txt#service_workers_and_shared_workers_support)
- [New adapter information attributes](https://developer.chrome.com/blog/new-in-webgpu-124.md.txt#new_adapter_information_attributes)
- [Bug fixes](https://developer.chrome.com/blog/new-in-webgpu-124.md.txt#bug_fixes)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-124.md.txt#dawn_updates)

### Chrome 123

- [DP4a built-in functions support in WGSL](https://developer.chrome.com/blog/new-in-webgpu-123.md.txt#dp4a_built-in_functions_support_in_wgsl)
- [Unrestricted pointer parameters in WGSL](https://developer.chrome.com/blog/new-in-webgpu-123.md.txt#unrestricted_pointer_parameters_in_wgsl)
- [Syntax sugar for dereferencing composites in WGSL](https://developer.chrome.com/blog/new-in-webgpu-123.md.txt#syntax_sugar_for_dereferencing_composites_in_wgsl)
- [Separate read-only state for stencil and depth aspects](https://developer.chrome.com/blog/new-in-webgpu-123.md.txt#separate_read-only_state_for_stencil_and_depth_aspects)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-123.md.txt#dawn_updates)

### Chrome 122

- [Expand reach with compatibility mode (feature in development)](https://developer.chrome.com/blog/new-in-webgpu-122.md.txt#expand_reach_with_compatibility_mode_feature_in_development)
- [Increase maxVertexAttributes limit](https://developer.chrome.com/blog/new-in-webgpu-122.md.txt#increase_maxvertexattributes_limit)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-122.md.txt#dawn_updates)

### Chrome 121

- [Support WebGPU on Android](https://developer.chrome.com/blog/new-in-webgpu-121.md.txt#support-webgpu-on-android)
- [Use DXC instead of FXC for shader compilation on Windows](https://developer.chrome.com/blog/new-in-webgpu-121.md.txt#use_dxc_instead_of_fxc_for_shader_compilation_on_windows)
- [Timestamp queries in compute and render passes](https://developer.chrome.com/blog/new-in-webgpu-121.md.txt#timestamp_queries_in_compute_and_render_passes)
- [Default entry points to shader modules](https://developer.chrome.com/blog/new-in-webgpu-121.md.txt#default_entry_points_to_shader_modules)
- [Support display-p3 as GPUExternalTexture color space](https://developer.chrome.com/blog/new-in-webgpu-121.md.txt#support_display-p3_as_gpuexternaltexture_color_space)
- [Memory heaps info](https://developer.chrome.com/blog/new-in-webgpu-121.md.txt#memory_heaps_info)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-121.md.txt#dawn_updates)

### Chrome 120

- [Support for 16-bit floating-point values in WGSL](https://developer.chrome.com/blog/new-in-webgpu-120.md.txt#support_for_16-bit_floating-point_values_in_wgsl)
- [Push the limits](https://developer.chrome.com/blog/new-in-webgpu-120.md.txt#push_the_limits)
- [Changes to depth-stencil state](https://developer.chrome.com/blog/new-in-webgpu-120.md.txt#changes_to_depth-stencil_state)
- [Adapter information updates](https://developer.chrome.com/blog/new-in-webgpu-120.md.txt#adapter_information_updates)
- [Timestamp queries quantization](https://developer.chrome.com/blog/new-in-webgpu-120.md.txt#timestamp_queries_quantization)
- [Spring-cleaning features](https://developer.chrome.com/blog/new-in-webgpu-120.md.txt#spring-cleaning_features)

### Chrome 119

- [Filterable 32-bit float textures](https://developer.chrome.com/blog/new-in-webgpu-119.md.txt#filterable_32-bit_float_textures)
- [unorm10-10-10-2 vertex format](https://developer.chrome.com/blog/new-in-webgpu-119.md.txt#unorm10-10-10-2_vertex_format)
- [rgb10a2uint texture format](https://developer.chrome.com/blog/new-in-webgpu-119.md.txt#rgb10a2uint_texture_format)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-119.md.txt#dawn_updates)

### Chrome 118

- [HTMLImageElement and ImageData support in `copyExternalImageToTexture()`](https://developer.chrome.com/blog/new-in-webgpu-118.md.txt#htmlimageelement_and_imagedata_support_in_copyexternalimagetotexture)
- [Experimental support for read-write and read-only storage texture](https://developer.chrome.com/blog/new-in-webgpu-118.md.txt#experimental_support_for_read-write_and_read-only_storage_texture)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-118.md.txt#dawn_updates)

### Chrome 117

- [Unset vertex buffer](https://developer.chrome.com/blog/new-in-webgpu-117.md.txt#unset_vertex_buffer)
- [Unset bind group](https://developer.chrome.com/blog/new-in-webgpu-117.md.txt#unset_bind_group)
- [Silence errors from async pipeline creation when device is lost](https://developer.chrome.com/blog/new-in-webgpu-117.md.txt#silence_errors_from_async_pipeline_creation_when_device_is_lost)
- [SPIR-V shader module creation updates](https://developer.chrome.com/blog/new-in-webgpu-117.md.txt#spir-v_shader_module_creation_updates)
- [Improving developer experience](https://developer.chrome.com/blog/new-in-webgpu-117.md.txt#improving_developer_experience)
- [Caching pipelines with automatically generated layout](https://developer.chrome.com/blog/new-in-webgpu-117.md.txt#caching_pipelines_with_automatically_generated_layout)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-117.md.txt#dawn_updates)

### Chrome 116

- [WebCodecs integration](https://developer.chrome.com/blog/new-in-webgpu-116.md.txt#webcodecs_integration)
- [Lost device returned by GPUAdapter `requestDevice()`](https://developer.chrome.com/blog/new-in-webgpu-116.md.txt#lost_device_returned_by_gpuadapter_requestdevice)
- [Keep video playback smooth if `importExternalTexture()` is called](https://developer.chrome.com/blog/new-in-webgpu-116.md.txt#keep_video_playback_smooth_if_importexternaltexture_is_called)
- [Spec conformance](https://developer.chrome.com/blog/new-in-webgpu-116.md.txt#spec_conformance)
- [Improving developer experience](https://developer.chrome.com/blog/new-in-webgpu-116.md.txt#improving_developer_experience)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-116.md.txt#dawn_updates)

### Chrome 115

- [Supported WGSL language extensions](https://developer.chrome.com/blog/new-in-webgpu-115.md.txt#supported_wgsl_language_extensions)
- [Experimental support for Direct3D 11](https://developer.chrome.com/blog/new-in-webgpu-115.md.txt#experimental_support_for_direct3d_11)
- [Get discrete GPU by default on AC power](https://developer.chrome.com/blog/new-in-webgpu-115.md.txt#get_discrete_gpu_by_default_on_ac_power)
- [Improving developer experience](https://developer.chrome.com/blog/new-in-webgpu-115.md.txt#improving_developer_experience)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-115.md.txt#dawn_updates)

### Chrome 114

- [Optimize JavaScript](https://developer.chrome.com/blog/new-in-webgpu-114.md.txt#optimize_javascript)
- [getCurrentTexture() on unconfigured canvas throws InvalidStateError](https://developer.chrome.com/blog/new-in-webgpu-114.md.txt#getcurrenttexture_on_unconfigured_canvas_throws_invalidstateerror)
- [WGSL updates](https://developer.chrome.com/blog/new-in-webgpu-114.md.txt#wgsl_updates)
- [Dawn updates](https://developer.chrome.com/blog/new-in-webgpu-114.md.txt#dawn_updates)

### Chrome 113

- [Use WebCodecs VideoFrame source in `importExternalTexture()`](https://developer.chrome.com/blog/new-in-webgpu-113.md.txt#use_webcodecs_videoframe_source_in_importexternaltexture)