Add ExecuTorch WebGPU program for whisper-base

README.md

@@ -16,14 +16,14 @@ tags:
 This repository contains an [ExecuTorch](https://github.com/pytorch/executorch) conversion of
 [openai/whisper-base](https://huggingface.co/openai/whisper-base) for the ExecuTorch **WebGPU** backend, so the model runs entirely in a web browser on
 the visitor's GPU. It powers the corresponding demo in the
-[ExecuTorch Playground](https://executorch.ai/playground/webgpu/).
+ExecuTorch WebGPU Playground (launching soon on [executorch.ai](https://executorch.ai)).
 
 - **Base model:** [openai/whisper-base](https://huggingface.co/openai/whisper-base)
 - **Format:** ExecuTorch `.pte` programs delegated to the WebGPU backend
 - **Quantization:** int8 token embeddings; other graphs as exported
 - **Total size:** 158.3 MB
 
-> **Note:** The Hugging Face model card lists Apache-2.0; the original code release is MIT. Confirm with legal which applies to redistributed weights.
+> **Note:** OpenAI released Whisper's code under MIT; the weights are distributed under Apache-2.0, as on the upstream Hugging Face model card.
 
 ## Files
 
@@ -46,14 +46,13 @@ Full checksums are in `manifest.json`. The playground verifies them after downlo
 
 These programs need the ExecuTorch WebGPU runtime compiled to WebAssembly
 ([backends/webgpu](https://github.com/pytorch/executorch/tree/main/backends/webgpu)). The easiest way to
-try the model is the [ExecuTorch Playground](https://executorch.ai/playground/webgpu/), which downloads these
-files once, caches them in the browser, and runs everything locally.
+try the model is the ExecuTorch WebGPU Playground (launching soon on [executorch.ai](https://executorch.ai)),
+which downloads these files once, caches them in the browser, and runs everything locally.
 
 ## Provenance
 
-- Converted from the upstream checkpoint with ExecuTorch's WebGPU export path.
-- Upstream revision: _to be recorded at export time_.
-- Export recipe: _to be published with the ExecuTorch release that ships the WebGPU runtime_.
+- Converted from the upstream checkpoint by the ExecuTorch team with the WebGPU export path.
+- A reproducible export recipe will be published alongside the open-source ExecuTorch WebGPU runtime.
 
 ## License