Feature/js sdk responses api

[MaanavD]

This pull request introduces comprehensive support for audio transcription (speech-to-text) using Whisper models in the Foundry Local SDK, alongside unified chat and audio capabilities. It adds new documentation, sample apps, and code examples demonstrating how to use both chat and audio features through a single SDK, highlighting automatic hardware acceleration and simplified integration. The updates make it easier for developers to build applications that combine text generation and speech-to-text without needing separate runtimes or hardware detection code.

Documentation and SDK feature updates:

• Updated README.md and docs/README.md to clarify that Foundry Local now supports both chat (text generation) and audio transcription (speech-to-text) in a single runtime, with automatic hardware acceleration and unified APIs. Added tables listing supported tasks, model aliases, and API usage. [1] [2]
• Expanded sample listings in documentation to include new JavaScript samples for chat, audio transcription, tool calling, and combined chat+audio apps. [1] [2]

New sample applications and examples:

• Added samples/js/audio-transcription-foundry-local sample, including a detailed README, package.json, and implementation in src/app.js, showing how to load Whisper models and transcribe audio files (standard and streaming modes). [1] [2] [3]
• Added samples/js/chat-and-audio-foundry-local sample, with README, package.json, and implementation in src/app.js, demonstrating unified management of chat and audio models, transcribing audio, and analyzing the transcription with a chat model. [1] [2] [3]

SDK example enhancements:

• Introduced a new TypeScript example audio-transcription.ts in sdk_v2/js/examples, showcasing programmatic use of Whisper models for audio transcription, including model discovery, download, loading, and both standard and streaming transcription.
• Added responses.ts example to sdk_v2/js/examples, demonstrating the Responses API for text generation, streaming, multi-turn conversations, tool calling, and response management.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
foundry-local	Ready	Preview, Comment	Mar 27, 2026 9:16pm

[Copilot]

Pull request overview

This PR expands the Foundry Local JavaScript developer experience by adding a new audio transcription (Whisper) sample and updating top-level documentation to show unified chat + audio usage via the JS SDK.

Changes:

• Added a JS sample app demonstrating Whisper model download/load and transcription.
• Added sample-scoped documentation and npm manifest for running the new sample.
• Updated the root README with richer JS examples for chat, audio transcription, and combined workflows.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.

File	Description
samples/js/audio-transcription-foundry-local/src/app.js	New sample implementation for standard + streaming audio transcription.
samples/js/audio-transcription-foundry-local/package.json	Adds a runnable npm package manifest for the sample.
samples/js/audio-transcription-foundry-local/README.md	Adds setup/run instructions and explains what the sample demonstrates.
README.md	Updates documentation examples for chat, audio transcription, and combined usage.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

This sample has a package.json with foundry-local-sdk already listed as a dependency, but the README instructs npm install foundry-local-sdk. That can leave users without the locked dependency tree for the sample; prefer npm install (then npm start / node src/app.js) from the sample directory.

Suggested change

	Install the Foundry Local SDK:
	```bash
	npm install foundry-local-sdk
	From this sample's directory, install dependencies:
	```bash
	npm install

[Copilot]

AudioClient.transcribeStreaming in the JS SDK returns an AsyncIterable and does not take a callback parameter. Calling it as await audioClient.transcribeStreaming(audioFilePath, (chunk) => ...) will throw/compile-fail; update the sample to iterate the returned stream (e.g., with for await ... of) and write each chunk.

Suggested change

	await audioClient.transcribeStreaming(audioFilePath, (chunk) => {
	process.stdout.write(chunk.text);
	});
	const stream = await audioClient.transcribeStreaming(audioFilePath);
	for await (const chunk of stream) {
	process.stdout.write(chunk.text);
	}

[Copilot]

Using the "latest" npm dist-tag makes this sample non-reproducible and can break unexpectedly when a new major version is published. Prefer a semver range (e.g. ^<current version>) or a workspace/file reference so the sample remains stable over time.

Suggested change

	"foundry-local-sdk": "latest"
	"foundry-local-sdk": "^1.0.0"