Update langchain4j monorepo to v1.8.0

[renovate]

This PR contains the following updates:

Package	Change	Age	Confidence
dev.langchain4j:langchain4j-bom (source)	1.1.0 -> 1.8.0
dev.langchain4j:langchain4j-core (source)	1.1.0 -> 1.8.0

---

Release Notes

langchain4j/langchain4j (dev.langchain4j:langchain4j-bom)

v1.8.0: and 1.8.0-beta15

Compare Source

Notable Changes

• Reflect latest MCP registry changes by @​jmartisk in #​3888
• Support tool name mapping for MCP clients by @​jmartisk in #​3799
• Allow the output guardrail to rewrite a complete AiMessage by @​mariofusco in #​3868
• Allow accessing AgenticScope from within @​Tool-annotated method by @​mariofusco in #​3785
• Allow passing custom attributes from RAG pipeline to ChatMemoryStore by @​dliubarskyi in #​3898
• Streaming Cancellation by @​dliubarskyi in #​3910

Other Changes

• docs: update home page to display features in card view by @​vijaypatidar in #​3809
• Add markdown & yaml parsers to bom by @​vesense in #​3814
• test: improve PojoListOutputParser test coverage for edge cases by @​alxkm in #​3805
• Split nightly workflows into multiple jobs by @​dliubarskyi in #​3822
• OpenAiStreamingChatModel: expose raw HTTP response by @​dliubarskyi in #​3828
• test: Add comprehensive test coverage for JSONMetadataHandler by @​alxkm in #​3823
• fix typo by @​jujn in #​3824
• Avoid returing a null AiMessage from Gemini by @​loicmathieu in #​3825
• Allow configuring the logger for logging MCP events by @​loicmathieu in #​3803
• Fix guardrails result after a successful reprompt by @​mariofusco in #​3820
• fix: the Builder exposes a logger setter but SseSubscriber still constructs its own Logger internally. by @​NeatGuyCoding in #​3835
• test: Add comprehensive test coverage for Vertex AI ContentsMapper by @​alxkm in #​3831
• Release automation by @​dliubarskyi in #​3838
• feat: Added support to auto reinitialize session if not found by @​vijaypatidar in #​3833
• Automate updating versions in documentation by @​dliubarskyi in #​3845
• Automate updating versions in documentation by @​dliubarskyi in #​3847
• test: enhance AiService chat memory configuration coverage by @​alxkm in #​3839
• Fix markdown tests package by @​vesense in #​3844
• docs: Add YugabyteDB embedding store documentation by @​krishna-yb in #​3852
• Azure CosmosDB No SQL: Fixing bugs by @​aayush3011 in #​3813
• Use LocalDateTime evaluated as UTC in the MCP registry client by @​jmartisk in #​3865
• Fix typos in WorkflowAgentsIT by @​vesense in #​3859
• Fix pure-agentic example by @​vesense in #​3860
• feat: I think the Document class should be public so that users can i… by @​GuXiangFly in #​3857
• docs(search-engine): DuckDuckGo integration by @​Malaydewangan09 in #​3862
• Add immutability and validation tests for DefaultQueryRouter by @​alxkm in #​3863
• Fix CI build failure due to langchain4j-docu-chatbot-updater license compliance error by @​vesense in #​3861
• docs: add OpenTelemetry GenAI instrumentation guide by @​dineshkumarkummara in #​3842
• test: Add validation and instance identity tests for ClassInstanceLoader by @​alxkm in #​3870
• docs: Add mariadb to embedding store list by @​vesense in #​3873
• Update the watsonx.ai SDK to the latest version by @​andreadimaio in #​3812
• Allow AiServices to return void by @​mariofusco in #​3874
• Renamed outputName to outputKey and updated docs by @​vijaypatidar in #​3877
• Retain description of anyOf elements from the MCP schema by @​jmartisk in #​3886
• docs: Remove examples from YugabyteDB documentation by @​krishna-yb in #​3875
• docs: Update chat memory docs by @​vesense in #​3876
• test: Add null key validation and empty state tests for InvocationParameters by @​alxkm in #​3881
• docs: add open-ai-official model to list by @​vesense in #​3882
• feat(open-ai): Add support to set query parameters to an OpenAI request by @​maff in #​3880
• test: Add validation tests and refactor AiServicesBuilderTest by @​alxkm in #​3891
• docs: update numbers in intro by @​vesense in #​3893
• chore(deps): Update Apache Tika to version 3.2.3 by @​maff in #​3885
• fix: thinking disabled should set thinking to null, toString should not expose secret by @​NeatGuyCoding in #​3892
• docs: update rag docs by @​vesense in #​3899
• Explicitly for the ResponseFormat instead of checking whether it's null when validating a ChatRequest for any Anthropic model by @​Bram-- in #​3900
• Fix typos in observability doc by @​jeanbisutti in #​3902
• test: add edge case coverage for JsonTestUtils.jsonify() by @​alxkm in #​3905
• feat: add support for base64 images using OpenAi Azure by @​mathieu-stennier in #​3909
• test: Add comprehensive test coverage for ParsingUtils by @​alxkm in #​3913
• Fix ConditionalInvocationHandler name by @​vesense in #​3922
• Add support for validation of content types in Ollama models and respective unit tests by @​nwijngaards in #​3923
• Extract a Base Gemini Chat Model builder by @​Bram-- in #​3917
• Enhance failure and fatal guardrail outcome descriptions by @​edeandrea in #​3919
• docs: update numbers in readme by @​vesense in #​3920
• test: Add comprehensive test coverage for TypeUtils by @​alxkm in #​3924
• test: Add missing test coverage for OpenAiUtils thinking content and edge cases by @​alxkm in #​3931
• Make GoogleAiGeminiChatModel unit-testable by @​Bram-- in #​3927
• Add tool execution request id for Gemini model by @​ChiaoGeek in #​3557
• Close resources in Utils.readBytes by @​jeanbisutti in #​3934
• test: add missing test coverage for ParsingUtils collection parsing by @​alxkm in #​3935
• Remove model from GeminiGenerateContentRequest by @​Bram-- in #​3936
• Fix NPE with OpenAiChatRequestParameters by @​jeanbisutti in #​3940

New Contributors

• @​jujn made their first contribution in #​3824
• @​krishna-yb made their first contribution in #​3852
• @​GuXiangFly made their first contribution in #​3857
• @​Malaydewangan09 made their first contribution in #​3862
• @​dineshkumarkummara made their first contribution in #​3842
• @​maff made their first contribution in #​3880
• @​Bram-- made their first contribution in #​3900
• @​jeanbisutti made their first contribution in #​3902
• @​mathieu-stennier made their first contribution in #​3909
• @​nwijngaards made their first contribution in #​3923

Full Changelog: langchain4j/langchain4j@1.7.1...1.8.0

v1.7.1: and 1.7.1-beta14

Notable Changes

• Agentic: allow to create agents based on classes by @​mariofusco in #​3786
• ChromaDB: support API V2 by @​vijaypatidar and @​ehsavoie in #​3779
• MCP: adjust to recent registry API changes by @​jmartisk in #​3793
• OpenAI Official: upgrade to SDK v4.0.0 by @​jdubois in #​3795
• Anthropic: allow setting custom parameters by @​dliubarskyi in #​3797
• Azure OpenAI: support maxCompletionTokens parameter @​dliubarskyi in faa12fa
• Hugging Face: deprecate HuggingFaceChatModel and HuggingFaceLanguageModel by @​dliubarskyi in #​3319

New Integrations

• Docker MCP transport by @​loicmathieu in #​3513
• Adding Oracle Document Loader implementation by @​psilberk in #​3114
• Add GPU Support to LangChain4j via GPULlama3.java Powered by TornadoVM by @​mikepapadim in #​3654
• Introduce yaml document parser by @​vesense in #​3748

Other Changes

• add logos and logo license by @​LizeRaes in #​3739
• test: Add validation and edge case tests for GoogleAI, IsNotEqualTo filter, and SQL builder by @​alxkm in #​3766
• Improve AgenticScope json serialization test by @​mariofusco in #​3777
• fix(gemini): Fix Google Gemini API field naming to use camelCase by @​minguncle in #​3567
• Support anyOf and null JSON schema elements in googleai.SchemaMapper by @​Zalewa in #​3781
• test: add comprehensive test coverage for InvocationParameters by @​alxkm in #​3787
• Disable flaky test in agentic module by @​mariofusco in #​3790
• test: enhance test coverage for IsEqualTo filter and Tablestore utilities by @​alxkm in #​3791
• fix: TokenStream should not require onPartialResponse() if onCompleteResponse() is used by @​vorburger in #​3792
• test: enhance PojoSetOutputParser test coverage by @​alxkm in #​3798

New Contributors

• @​vorburger made their first contribution in #​3792
• @​mikepapadim made their first contribution in #​3654

Full Changelog: langchain4j/langchain4j@1.6.0...1.7.1

v1.6.0: and 1.6.0-beta12

Compare Source

Notable Changes

• Agentic
  • Ensure that all pending async agent invocations are completed before ending the root call by @​mariofusco in #​3695
  • Call output function when a loop terminates with the exit condition by @​mariofusco in #​3696
  • Allow to mix agentic declarative and programmatic API by @​mariofusco in #​3699
  • Use unique names for agents inside the same agentic system by @​mariofusco in #​3703
  • Fix agentic context configuration propagation in declarative API by @​mariofusco in #​3709
  • Fix A2A client after introduction of agents' unique names by @​mariofusco in #​3711
  • Upgrading A2A SDK to 0.3.0.Beta1 by @​ehsavoie in #​3710
  • Tweak supervisor prompt by @​mariofusco in #​3726
  • Improve agentic loop exit condition by @​mariofusco in #​3713
  • Introduce agents listeners by @​mariofusco in #​3718
  • Add declarative HumanInTheLoop by @​mariofusco in #​3735
  • Allow to configure ChatMemoryProvider on SupervisorAgent by @​mariofusco in #​3740
  • Add declarative API for A2A client by @​mariofusco in #​3755
• MCP
  • Adjust to recent changes in the official MCP registry by @​jmartisk in #​3723
  • Rework MCP client to support structured content results by @​jmartisk in #​3733
  • When cancelling a tool, send the requestId as a number by @​jmartisk in #​3707
  • Make the parsing of SSE events more robust by @​jmartisk in #​3724
• AI Services
  • Backport observability from Quarkus LangChain4j by @​edeandrea in #​3679
  • Option to pass extra parameters into and between Tool/ToolProvider/RAG components by @​dliubarskyi in #​3569

New Integrations

• Azure Container Apps Dynamic Sessions by @​bbenz in #​3266
• Markdown document parser by @​vesense in #​3732

Other Changes

• test: Add comprehensive test coverage for AiServices, Metadata, and StringSetOutputParsert by @​alxkm in #​3690
• Add image URL support for Anthropic Claude vision models(#​3691) by @​henrydo300 in #​3692
• Updates couchbase client to 3.9.1 by @​chedim in #​3702
• Refactor addAll method to validate ids list and ensure consistent lis… by @​harikrishna553 in #​3681
• test: Enhance test coverage for core component validation and edge cases by @​alxkm in #​3706
• Test for 'data:' inside a SSE event by @​jmartisk in #​3725
• test: Enhance test coverage for listener error handling, content aggregation, and enum parsing by @​alxkm in #​3719
• Fix typo in 'Anthropic' reference by @​vesense in #​3731
• test: Add validation and edge case test coverage for core request/response handling by @​alxkm in #​3727
• test: Add validation test coverage for query routing and configuration components by @​alxkm in #​3737
• Add ensureNotNull validation and unit tests for McpResource and McpRe… by @​harikrishna553 in #​3705
• feat: Added batching support for bedrock cohere embedding model by @​vijaypatidar in #​3730
• Follow-up to #​3685: migrate nullability annotations to JSpecify by @​kushal-dixit in #​3694
• Add Javadocs to declarative API annotations by @​mariofusco in #​3747
• Test against Java 25 by @​dliubarskyi in #​3751
• test: Add comprehensive test coverage for ChatResponse, ChromaMetadataFilter, and DefaultQueryRouter by @​alxkm in #​3750
• docs: add custom-parameters option for OpenAI by @​JungahGoak in #​3749
• test: Add critical validation and edge case tests for AiServices, StreamingChatModelMock, and TablestoreEmbeddingStore by @​alxkm in #​3759

New Contributors

• @​henrydo300 made their first contribution in #​3692
• @​vijaypatidar made their first contribution in #​3730
• @​JungahGoak made their first contribution in #​3749

Full Changelog: langchain4j/langchain4j@1.5.0...1.6.0

v1.5.0: and 1.5.0-beta11

Compare Source

Notable Changes

• MCP: initial registry client implementation by @​jmartisk in #​3684
• Agentic: asynchronous agents by @​mariofusco in #​3643
• Agentic: missing tools features by @​mariofusco in #​3660
• Google AI Gemini: support image outputs by @​minguncle in #​3641
• AWS Bedrock: support prompt caching by @​valimikayilov in #​3664

Other Changes

• Add missing StyleEditor implementation in Agents tutorial by @​harikrishna553 in #​3611
• Fix #​3650 by @​dliubarskyi in #​3655
• Refactor addInternal to remove redundant HashMap initialization by @​harikrishna553 in #​3618
• Fix invocation of output function in supervisor agent when max agents… by @​mariofusco in #​3623
• Propagate declaratively supplied ChatModel from main agent to subagents by @​mariofusco in #​3625
• Fix json serialization of AgenticScope by @​mariofusco in #​3656
• #​3628: Use received description for visited json object schema by @​jordibisbal8 in #​3661
• Update the watsonx.ai SDK to the latest version by @​andreadimaio in #​3644
• Document support A2A response types by @​mbroz2 in #​3640
• test: Enhance test coverage for ConversationalChain and ConversationalRetrievalChain by @​alxkm in #​3610
• fix(deps): update dependency software.amazon.awssdk:bom to v2.33.5 by @​renovate[bot] in #​3435
• fix(deps): update dependency org.mockito:mockito-bom to v5.19.0 by @​renovate[bot] in #​3483
• fix(deps): update dependency ch.qos.logback:logback-classic to v1.5.18 by @​renovate[bot] in #​3436
• chore(deps): update dependency org.wiremock:wiremock-standalone to v3.13.1 by @​renovate[bot] in #​3486
• Fix GoogleAiGeminiTokenCountEstimator sending empty request for system message by @​linnaung in #​3616
• Allow configuring the logger for logging requests and responses by @​loicmathieu in #​3612
• fix: wait for the tool-list update to finish by @​jiangsier-xyz in #​3630
• Refactor registerContext for clarity and safer handling of chat messages by @​harikrishna553 in #​3631
• added requirements for common ITs by @​dliubarskyi in #​3665
• Allow to create an agentic system made of a single agent using the declarative API by @​mariofusco in #​3666
• feat(anthropic): propagate user id in Anthropic models by @​maxlepikhin in #​3670
• Add callback allowing to configure agents created through the declarative API by @​mariofusco in #​3672
• test: Add comprehensive test coverage for AwsDocumentConverter, ChatRequest, and Content classes by @​alxkm in #​3673
• Record agent messages in the AgenticScope even without a ChatMemory by @​mariofusco in #​3675
• Clarify the documentation by using static imports by @​jdubois in #​3677
• feat(Anthropic user id): fixed AnthropicUserIdIT by @​maxlepikhin in #​3676
• test: Add comprehensive test coverage for OpenAiChatRequestParameters, CustomMimeTypesFileTypeDetector, and Query classess by @​alxkm in #​3683
• Introduce AiServiceContextFactory by @​mariofusco in #​3687
• Anthropic: added the ability to turn off parallel tool calls and improved the toolChoice implementation by @​Claudio-code in #​3401
• Fix null handling in JinaEmbeddingModel response and mask apiKey in t… by @​harikrishna553 in #​3680
• Remove Lombok annotations from remaining code and replace with standard java checks and implementations. by @​kushal-dixit in #​3685

New Contributors

• @​jordibisbal8 made their first contribution in #​3661
• @​mbroz2 made their first contribution in #​3640
• @​maxlepikhin made their first contribution in #​3670
• @​minguncle made their first contribution in #​3641
• @​valimikayilov made their first contribution in #​3664
• @​kushal-dixit made their first contribution in #​3685

Full Changelog: langchain4j/langchain4j@1.4.0...1.5.0

v1.4.0: and 1.4.0-beta10

Compare Source

Notable Changes

• Agentic: add missing features to declarative API by @​mariofusco in #​3587
• Agentic: allow to manipulate the supervisor output before returning it by @​mariofusco in #​3602
• MCP: Streamable HTTP transport by @​jmartisk in #​3534
• AI Service: option to execute tools concurrently by @​dliubarskyi in #​3512
• AI Service: support immediate/direct return from tools by @​mariofusco in #​3069
• AI Service: introduce tool arguments and tool execution error handlers by @​dliubarskyi in #​3536
• AI Serivce: multimodality support by @​miracle-bean in #​3420
• Structured Outputs: introduce a raw JSON-schema element by @​ivy-rew in #​3374

New Integrations

• Integrate watsonx.ai module by @​andreadimaio in #​3423
• Azure OpenAI: support audio transcription model by @​SandraAhlgrimm in #​2110
• Anthropic: add support for counting tokens by @​HamzaElzarw-2022 in #​3381

Breaking Changes

• AI Service: introduce tool arguments and tool execution error handlers by @​dliubarskyi in #​3536
• Add missing features to agentic declarative API by @​mariofusco in #​3587
• Use a pool of virtual threads as internal default executor by @​mariofusco in #​3541

Other Changes

• add mariofusco as developer of the agentic modules by @​mariofusco in #​3502
• minor fix to agents tutorial by @​mariofusco in #​3504
• Improve OrganicSearchResult Javadoc, Constructor Clarity, and Metadat… by @​harikrishna553 in #​3505
• fix in Azure Document Loader Dependency: update github-api to 1.329 to resolve commons-io security vulnerability by @​SandraAhlgrimm in #​3506
• Remove unwanted links from images in tutorials by @​mariofusco in #​3517
• Allow using the MCP ToolExecutor outside of the McpToolProvider by @​loicmathieu in #​3494
• Fix building of agentic declarative parallel workflow by @​mariofusco in #​3530
• fix Ollama agentic tests by @​donbourne in #​3539
• minor wording fixes by @​donbourne in #​3518
• Allow passing a system message to the agent via the builder by @​loicmathieu in #​3533
• Fix: Update Vespa version to address Jetty HTTP security vulnerability by @​SandraAhlgrimm in #​3549
• #​3550 Fix typo on field name variables by @​cunningt in #​3551
• fix:[#​3552] LoggingHttpClient should not use the DefaultServerSentEve… by @​codexvn in #​3554
• Allow passing null memory or memory provider by @​loicmathieu in #​3525
• test: Add comprehensive edge case tests for Content class by @​alxkm in #​3521
• chore(deps): update dependency org.apache.maven.plugins:maven-javadoc-plugin to v3.11.3 by @​renovate[bot] in #​3490
• Inject business/domain context into the planning phase by @​mnhnam in #​3527
• Fix a NPE that occurs if there is no parsed SSE by @​edeandrea in #​3563
• Fix code formatting in Azure OpenAI and Core modules by @​SandraAhlgrimm in #​3577
• [minor fix] Remove unnecessary imports by @​vesense in #​3279
• Use a pool of virtual threads as internal default executor by @​mariofusco in #​3541
• Add utility methods to find SystemMessage instances in chat messages by @​agoncal in #​3582
• Allows to programmatically define the name and description of an agent by @​mariofusco in #​3594
• Upgrade to OpenAI SDK Java 3.1.2 by @​jdubois in #​3593
• Docs: fix typo in pom type configuration tag by @​Silwings-git in #​3597
• #​3588: Handle IllegalArgumentException during URI.create by @​msampietro in #​3589
• Allow to evict or completely disable the tool list cache in DefaultMcpClient by @​Zalewa in #​3578
• Feature/mcp sse add header by @​Y7000-save-me in #​3570
• Allow configuring the logger for logging requests and responses by @​loicmathieu in #​3561
• Update playwright version to 1.54.0 by @​vesense in #​3584
• added neo4j follow-up query docs by @​vga91 in #​3583
• Improve test coverage for logical filter classes (And, Or, Not) by @​alxkm in #​3564
• Add unit tests for default map name handling and fix Javadoc in Coher… by @​harikrishna553 in #​3507
• Allow setting the JSON Schema in the OpenAI ChatModel by @​loicmathieu in #​3598
• Add logRequests and logResponses to Google Gemini AI by @​loicmathieu in #​3565
• Improve validation and Javadoc for EmbeddingModelTextClassifier by @​harikrishna553 in #​3466
• Refactor equals methods to use pattern matching and eliminate raw typ… by @​harikrishna553 in #​3462
• fix (google-ai-gemini): avoid NPE when Gemini filters content (parts == null) (#​2893) by @​linnaung in #​3603
• Allow configuring the logger for logging requests and responses by @​loicmathieu in #​3601
• Update watsonx.ai SDK and implement tokenization methods by @​andreadimaio in #​3605
• GH-3369 Support for encodingFormat in OpenAI's embedding model by @​sunyuhan1998 in #​3370
• Added the ability to specify the values of additional parameters in the creation of Milvus indexes by @​odysseaspenta in #​3464
• AI Service: fix implementation of methods declared in Object by @​dliubarskyi in #​3459
• Azure CosmosDB No SQL: Sharded DiskANN, Content Retriever Store and Memory Store by @​aayush3011 in #​3463
• Add custom retry options to Azure OpenAI models by @​canberk-oguz in #​3591
• Allow serializing null properties for Bedrock by @​jmartisk in #​3607
• Warn about non-JSON messages that MCP clients receive over stdio by @​jmartisk in #​3609
• Feat/logger provider 3 by @​loicmathieu in #​3608

New Contributors

• @​SandraAhlgrimm made their first contribution in #​3506
• @​loicmathieu made their first contribution in #​3494
• @​ivy-rew made their first contribution in #​3374
• @​donbourne made their first contribution in #​3539
• @​cunningt made their first contribution in #​3551
• @​codexvn made their first contribution in #​3554
• @​mnhnam made their first contribution in #​3527
• @​Silwings-git made their first contribution in #​3597
• @​msampietro made their first contribution in #​3589
• @​andreadimaio made their first contribution in #​3423
• @​Y7000-save-me made their first contribution in #​3570
• @​miracle-bean made their first contribution in #​3420
• @​HamzaElzarw-2022 made their first contribution in #​3381
• @​linnaung made their first contribution in #​3603
• @​odysseaspenta made their first contribution in #​3464

Full Changelog: langchain4j/langchain4j@1.3.0...1.4.0

v1.3.0: and 1.3.0-beta9

Compare Source

Announcements

We're excited to announce the release of two new experimental modules - langchain4j-agentic and langchain4j-agentic-a2a - which introduce a set of abstractions and utilities for building agentic, AI-infused applications. See the documentation for more information.

What's Changed

• Introduce new langchain4j-agentic module by @​mariofusco in #​3458
• Docs for Infinispan Embedded Store by @​karesti in #​3450
• [Fix] Incorrect join changes in Infinispan by @​karesti in #​3454
• Fix bug in output guardrails when maxRetries == 0 by @​edeandrea in #​3470
• Update dependency com.couchb

---

Configuration

📅 Schedule: Branch creation - At any time (no schedule defined), Automerge - At any time (no schedule defined).

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about these updates again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.