feat!: Support invoke with structured output in LangChain provider #970
+155 −31
Conversation
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
Contributor Author
| This PR is reliant on #969 being merged first. |
Contributor
| @launchdarkly/browser size report |
Contributor
| @launchdarkly/js-sdk-common size report |
Contributor
| @launchdarkly/js-client-sdk size report |
Contributor
| @launchdarkly/js-client-sdk-common size report |
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
packages/ai-providers/server-ai-langchain/__tests__/LangChainProvider.test.ts Show resolved Hide resolved
abarker-launchdarkly approved these changes Nov 4, 2025
jsonbailey changed the title 
Merged
jsonbailey added a commit that referenced this pull request Nov 6, 2025
🤖 I have created a release *beep* *boop* --- <details><summary>server-sdk-ai: 0.14.0</summary> ## [0.14.0](server-sdk-ai-v0.13.0...server-sdk-ai-v0.14.0) (2025-11-06) ### ⚠ BREAKING CHANGES * Removed deprecated Vercel methods ([#983](#983)) * Add support for real time judge evals ([#969](#969)) * AI Config defaults require the "enabled" attribute * Renamed LDAIAgentConfig to LDAIAgentConfigRequest for clarity * Renamed LDAIAgent to LDAIAgentConfig *note the previous use of this name * Renamed LDAIAgentDefault to LDAIAgentConfigDefault for clarity * Renamed LDAIDefaults to LDAICompletionConfigDefault for clarity ### Features * Add support for real time judge evals ([#969](#969)) ([6ecd9ab](6ecd9ab)) * Added createJudge method ([6ecd9ab](6ecd9ab)) * Added judgeConfig method to AI SDK to retrieve an AI Judge Config ([6ecd9ab](6ecd9ab)) * Added trackEvalScores method to config tracker ([6ecd9ab](6ecd9ab)) * Chat will evaluate responses with configured judges ([6ecd9ab](6ecd9ab)) * Include AI SDK version in tracking information ([#985](#985)) ([ef90564](ef90564)) * Removed deprecated Vercel methods ([#983](#983)) ([960a499](960a499)) ### Bug Fixes * AI Config defaults require the "enabled" attribute ([6ecd9ab](6ecd9ab)) * Renamed LDAIAgent to LDAIAgentConfig *note the previous use of this name ([6ecd9ab](6ecd9ab)) * Renamed LDAIAgentConfig to LDAIAgentConfigRequest for clarity ([6ecd9ab](6ecd9ab)) * Renamed LDAIAgentDefault to LDAIAgentConfigDefault for clarity ([6ecd9ab](6ecd9ab)) * Renamed LDAIDefaults to LDAICompletionConfigDefault for clarity ([6ecd9ab](6ecd9ab)) </details> <details><summary>server-sdk-ai-langchain: 0.3.0</summary> ## [0.3.0](server-sdk-ai-langchain-v0.2.0...server-sdk-ai-langchain-v0.3.0) (2025-11-06) ### ⚠ BREAKING CHANGES * Support invoke with structured output in LangChain provider ([#970](#970)) ### Features * Support invoke with structured output in LangChain provider ([#970](#970)) ([0427908](0427908)) ### Dependencies * The following workspace dependencies were updated * devDependencies * @launchdarkly/server-sdk-ai bumped from ^0.13.0 to ^0.14.0 * peerDependencies * @launchdarkly/server-sdk-ai bumped from ^0.12.2 to ^0.14.0 </details> <details><summary>server-sdk-ai-openai: 0.3.0</summary> ## [0.3.0](server-sdk-ai-openai-v0.2.0...server-sdk-ai-openai-v0.3.0) (2025-11-06) ### ⚠ BREAKING CHANGES * Support invoke with structured output in OpenAI provider ([#980](#980)) ### Features * Support invoke with structured output in OpenAI provider ([#980](#980)) ([515dbdf](515dbdf)) ### Dependencies * The following workspace dependencies were updated * devDependencies * @launchdarkly/server-sdk-ai bumped from ^0.13.0 to ^0.14.0 * peerDependencies * @launchdarkly/server-sdk-ai bumped from ^0.12.2 to ^0.14.0 </details> <details><summary>server-sdk-ai-vercel: 0.3.0</summary> ## [0.3.0](server-sdk-ai-vercel-v0.2.0...server-sdk-ai-vercel-v0.3.0) (2025-11-06) ### ⚠ BREAKING CHANGES * Support invoke with structured output in VercelAI provider ([#981](#981)) ### Features * Support invoke with structured output in VercelAI provider ([#981](#981)) ([d0cb41d](d0cb41d)) ### Dependencies * The following workspace dependencies were updated * devDependencies * @launchdarkly/server-sdk-ai bumped from ^0.13.0 to ^0.14.0 * peerDependencies * @launchdarkly/server-sdk-ai bumped from ^0.12.2 to ^0.14.0 </details> --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please). <!-- CURSOR_SUMMARY --> --- > [!NOTE] > Release server-ai 0.14.0 (judge evals, breaking renames/removals) and update LangChain/OpenAI/Vercel providers to 0.3.0 with structured output; refresh examples and manifests to new versions. > > - **SDK (`packages/sdk/server-ai`) — `0.14.0`** > - Adds real-time judge evaluations and related APIs (`createJudge`, `judgeConfig`, `trackEvalScores`); includes SDK version in tracking. > - Breaking: removes deprecated Vercel methods; requires `enabled` in AI Config defaults; renames several AI config types. > - **AI Providers — `0.3.0`** > - `@launchdarkly/server-sdk-ai-langchain`, `-openai`, `-vercel`: add structured output support for `invoke` (breaking changes). > - Bump peer/dev dependency on `@launchdarkly/server-sdk-ai` to `^0.14.0`. > - **Examples** > - Update example apps to use `@launchdarkly/server-sdk-ai@0.14.0` and provider packages `^0.3.0`. > - **Release metadata** > - Update `.release-please-manifest.json` with new versions. > > <sup>Written by [Cursor Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit 00cd808. This will update automatically on new commits. Configure [here](https://cursor.com/dashboard?tab=bugbot).</sup> <!-- /CURSOR_SUMMARY --> --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: jsonbailey <jbailey@launchdarkly.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
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.
Note
Introduces invokeStructuredModel for structured outputs and wraps invokeModel with try/catch to return consistent failure responses with logging; updates tests accordingly.
src/LangChainProvider.ts):invokeStructuredModel(messages, responseStructure)usingwithStructuredOutput(...), returning{ data, rawResponse, metrics }and handling errors with warnings and failure metrics.invokeModelin try/catch; on failure, log warning and return empty assistant message withsuccess=false.StructuredResponsetype.createLangChainModel(spreadparametersbeforemodelProvider).__tests__/LangChainProvider.test.ts):invokeModelerror path (logs and failure response).invokeStructuredModelsuccess and error behaviors.Written by Cursor Bugbot for commit 7560025. This will update automatically on new commits. Configure here.