Open
Conversation
…and SDK examples - Update README.md to prominently feature audio transcription (STT) alongside chat completions, including a Supported Tasks table, JS code examples for audio transcription and unified chat+audio, and updated Features section - Add samples/js/audio-transcription-foundry-local: standalone Whisper STT sample - Add samples/js/chat-and-audio-foundry-local: unified chat + audio sample demonstrating single FoundryLocalManager managing both model types - Add sdk_v2/js/examples/audio-transcription.ts: TypeScript audio example - Update docs/README.md with capabilities table and sample links Addresses the discoverability gap where LLMs and developers do not know Foundry Local supports audio transcription via Whisper models.
- Adopt upstream's restructured welcome section (key benefits list, collapsible SDK examples) while preserving audio transcription additions - Merge Supported Tasks table, audio transcription code example, and enhanced samples table descriptions into upstream's new format - Add Foundry Local Lab link from upstream to docs/README.md alongside our capabilities table and samples links - Drop standalone Features section (key points already in welcome benefits)
…tibility Add ResponsesClient to the JS SDK v2 with full CRUD support for the Responses API served by Foundry Local's embedded web service. New files: - src/openai/responsesClient.ts: HTTP-based client with SSE streaming - test/openai/responsesClient.test.ts: 35 tests (unit + integration) - examples/responses.ts: end-to-end usage examples Modified files: - src/types.ts: Responses API types (request, response, items, events) - src/index.ts: export ResponsesClient, ResponsesClientSettings, getOutputText - src/foundryLocalManager.ts: createResponsesClient() factory - src/imodel.ts: createResponsesClient(baseUrl) on IModel interface - src/model.ts, src/modelVariant.ts: delegation/implementation Key design decisions: - HTTP-based (fetch + SSE), not FFI, since no CoreInterop command exists - Factory on FoundryLocalManager (owns URL) + convenience on Model - Types scoped to neutron-server's supported feature set - Follows ChatClient patterns: settings serialization, validation, error chaining - getOutputText() helper matching OpenAI Python SDK's response.output_text
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
The regex /\/+$/ used to strip trailing slashes from baseUrl was flagged as a polynomial regular expression (ReDoS risk) by CodeQL. Replaced with a simple while/endsWith/slice loop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
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:
README.mdanddocs/README.mdto 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]New sample applications and examples:
samples/js/audio-transcription-foundry-localsample, including a detailed README,package.json, and implementation insrc/app.js, showing how to load Whisper models and transcribe audio files (standard and streaming modes). [1] [2] [3]samples/js/chat-and-audio-foundry-localsample, with README,package.json, and implementation insrc/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:
audio-transcription.tsinsdk_v2/js/examples, showcasing programmatic use of Whisper models for audio transcription, including model discovery, download, loading, and both standard and streaming transcription.responses.tsexample tosdk_v2/js/examples, demonstrating the Responses API for text generation, streaming, multi-turn conversations, tool calling, and response management.