0.16.0

What's Changed

• Add Ollama note in the advantages section by @geoand in #657
• Set a smaller segment-size for the easy-rag sample by @jmartisk in #661
• Allow creating and launching an OpenWebUI container via Dev UI by @iocanel in #600
• Support streaming in dev-ui chat by @iocanel in #590
• Reuse embeddings by @edeandrea in #666
• Avoid trying to retrieve the default ds if a named one should be used by @jmartisk in #668
• Update Quarkus version in samples by @geoand in #672
• Use top-level timeout property in samples by @geoand in #674
• Remove preceding and trailing json markers when converting from json by @geoand in #673
• Add secure-fraud-detection demo by @sberyozkin in #539
• Bump quarkus-neo4j.version from 4.0.0 to 4.1.0 by @dependabot in #669
• Update the email me a poem sample by @geoand in #678
• Implement logging for Chroma and attempt to fix the test by @jmartisk in #686
• (doc) small ollama typo by @laurentperez in #687
• Bump org.apache.maven.plugins:maven-failsafe-plugin from 3.2.5 to 3.3.0 by @dependabot in #688
• Dev UI: Allow RAG and tools together with streaming by @jmartisk in #689
• Add a link to the workshop into our docs by @jmartisk in #693
• Fix the nightly job by @jmartisk in #703
• Allow the use of the {response_schema} in the prompt by @andreadimaio in #699
• Allow the injection of TokenCountEstimator by @andreadimaio in #705
• Add domain config for Azure AI by @duchamk in #698
• Bump org.apache.maven.plugins:maven-surefire-plugin from 3.2.5 to 3.3.0 by @dependabot in #681
• Bump org.mvnpm.at.mvnpm:vaadin-webcomponents from 24.3.13 to 24.4.0 by @dependabot in #680
• Putting Generic Authentication flow to the Core by @csotiriou in #694
• Update pgvector-store.adoc - remove reference to Redis by @reconway in #711
• Hardcode neo4j image version to 5.20.0 for the tests for now by @jmartisk in #715
• Add a warning about missing -parameters flag for javac by @geoand in #713
• Redis embedding store: metadata filtering by @jmartisk in #714
• Bump org.wiremock:wiremock-standalone from 3.6.0 to 3.8.0 by @dependabot in #712
• Bump org.apache.maven.plugins:maven-surefire-plugin from 3.2.5 to 3.3.0 by @dependabot in #710
• Switch category of extensions to AI by @cescoffier in #704
• Bringing domain-name on par with the rest of the optional configurations by @csotiriou in #707
• Improve UI of the embedding store page by @jmartisk in #720
• Rename one of the OpenAiDevUIProcessors to avoid name clash by @jmartisk in #719
• Bump Quarkus version to latest by @geoand in #717
• Prevent false positive of javac -parameters by @geoand in #718
• Bump io.quarkiverse.wiremock:quarkus-wiremock-test from 1.3.2 to 1.3.3 by @dependabot in #677
• add stop sequence config for openAI chat config model by @sebastienblanc in #695
• Update to LangChain4j 0.32.0 by @geoand in #721

New Contributors

• @sberyozkin made their first contribution in #539
• @laurentperez made their first contribution in #687
• @duchamk made their first contribution in #698
• @reconway made their first contribution in #711

Full Changelog: 0.15.1...0.16.0

0.15.1

What's Changed

• Replace quarkus PGVectorEmbeddingStore with new Langchain4J PGVectorEmbeddingStore by @humcqc in #619
• Ollama Illegal unquoted character by @humcqc in #651
• Bump org.wiremock:wiremock-standalone from 3.5.4 to 3.6.0 by @dependabot in #615
• Don't fail in the DevUI because of WebSearchTool by @geoand in #653

Full Changelog: 0.15.0...0.15.1

0.15.0

What's Changed

• Fix bearer pattern for logging Azure requests by @geoand in #601
• Add default values for URLS of HuggingFace models by @geoand in #604
• Rename csv-chatbot to sql-chatbot by @jmartisk in #605
• Restructure the code of the repository by @geoand in #606
• Polish AiServicesRecorder by @geoand in #608
• Add an example of use of image description using GPT-4o by @geoand in #614
• Update to LangChain4j 0.31.0 by @geoand in #622
• Simplify OpenTelemetry instrumentation code by @geoand in #623
• Fix issue with reload when tools are new added by @geoand in #629
• Mistral listed twice in nav by @edeandrea in #641
• Remove erroneously added service file by @geoand in #645
• make vertexai.gemini timeout configurable by @exaV in #640
• Watch files configured in @SystemMessage or @UserMessage by @geoand in #639
• Bump org.assertj:assertj-core from 3.25.3 to 3.26.0 by @dependabot in #627
• Allow dynamic authorization headers for OpenAI + Azure OpenAI by @csotiriou in #646
• Introduce a way to configure tools per method of an AiService by @geoand in #613

Full Changelog: 0.14.1...0.15.0

0.14.2

What's Changed

• Fix bearer pattern for logging Azure requests by @geoand in #601
• Fix issue with reload when tools are new added by @geoand

Full Changelog: 0.14.1...0.14.2

0.14.1

What's Changed

• Refactor jokes jbang bot to be picocli based by @iocanel in #571
• Properly configure for embedding model by @csotiriou in #597
• Fix update sample version when releasing by @iocanel in #596
• Allow multi-deployment configuration for Azure OpenAI by @csotiriou in #598

Full Changelog: 0.14.0...0.14.1

0.14.0

What's Changed

• Fix configuration error in Gemini by @exaV in #558
• Upgrade to Quarkus Antora 0.0.4 - fix podman issues by @ppalaga in #533
• Bump org.mvnpm.at.mvnpm:vaadin-webcomponents from 24.3.11 to 24.3.12 by @dependabot in #553
• Bump io.quarkiverse.wiremock:quarkus-wiremock-test from 1.3.1 to 1.3.2 by @dependabot in #526
• Include instruction for docker users in email-me-a-poem readme by @iocanel in #562
• Drop uneeded MyChatModelSupplier from email-me-a-peom by @iocanel in #565
• Add minimal html UI for email me a peom by @iocanel in #564
• Fix distance fraud detection by @iocanel in #568
• Fix description of email-me-a-poem by @iocanel in #561
• Setup now ensures there are frauds to report by @iocanel in #566
• Add hints in the triage review page by @iocanel in #570
• Add documentation on using authentication headers with Ollama by @geoand in #573
• Fix the poem and triage samples by @cescoffier in #574
• When there is a single SystemMessage expose set it to dev ui. by @iocanel in #572
• Add missing guide link in extensions by @iocanel in #576
• Use vaadin progress bar in dev-ui embedding store by @iocanel in #575
• Introduce a new foundation for Dev Services like experience by @geoand in #557
• Apply model constraints to image ui combos by @iocanel in #579
• Improve email-me-a-poem ui by @iocanel in #582
• Bump org.mvnpm.at.mvnpm:vaadin-webcomponents from 24.3.12 to 24.3.13 by @dependabot in #583
• Update review triage sample by @geoand in #584
• Fix name in VertxAiGeminiRestApi by @geoand in #587
• Use GPT 4o in fraud detection sample by @geoand in #586
• Introduce preloading of models for Ollama by @geoand in #585
• Add docs about using Podman AI Lab by @geoand in #588
• Make token limit configuration for Ollama optional by @geoand in #589
• Introduce global timeout configuration by @geoand in #593
• Segregate memory of concurrent requests to different AI services by @geoand in #594

New Contributors

• @exaV made their first contribution in #558

Full Changelog: 0.13.1...0.14.0

0.13.1

What's Changed

• Send pretty printed JSON body by @geoand in #550
• Neo4j embedding store by @jmartisk in #548
• Bump io.sundr:sundr-maven-plugin from 0.103.1 to 0.200.0 by @dependabot in #527

Full Changelog: 0.13.0...0.13.1

0.13.0

What's Changed

• Bump org.apache.maven.plugins:maven-compiler-plugin from 3.12.1 to 3.13.0 by @dependabot in #516
• Ability to turn off ingestion with Easy RAG by @jmartisk in #518
• Introduce dependabot updates for samples by @geoand in #517
• Move BAM APIs version to 2024-04-15 by @andreadimaio in #520
• Add a Redis based chat memory store extension by @sebastienblanc in #429
• Improve a few things in websockets.adoc by @gsmet in #521
• Make sure the bot knows its name in samples by @geoand in #524
• Use @default literal instance by @geoand in #525
• Remove timestamps from chat messages for now by @jmartisk in #528
• Improve the Easy RAG documentation by @jmartisk in #530
• Introduce initial support for Google's Palm2 chat model by @geoand in #531
• Make Vertex AI module testable by @geoand in #532
• Introduce initial support for Google's Gemini chat model by @geoand in #534
• Properly catch JSON-RPC errors when adding embeddings by @jmartisk in #537
• Introduce global logging properties by @geoand in #536
• Add a little tip about quarkus.langchain4j.log-requests/responses by @jmartisk in #538
• feat: ✨ ollama streaming mode by @philippart-s in #522
• Use proper Ollama API for chats by @geoand in #543
• Add new parameters to the watsonx.ai chat model by @andreadimaio in #544
• Move the parsers out of the core module by @geoand in #546

Full Changelog: 0.12.1...0.13.0

0.12.1

What's Changed

• Add READMEs to all samples by @jmartisk in #499
• Improve user experience for samples by @geoand in #500
• Override the maven-compiler-plugin version in the samples by @jmartisk in #503
• JBang sample by @jmartisk in #504
• One forgotten compiler-plugin.version override by @jmartisk in #506
• Turn off built-in fault tolerance by default by @jmartisk in #507
• Update websockets.adoc by @gastaldi in #509
• Use pattern matching to simplify OpenAI REST code by @geoand in #502
• fix: 🐛 fix NPE if chunk is null for Mistral AI by @philippart-s in #510
• Bump org.wiremock:wiremock-standalone from 3.5.3 to 3.5.4 by @dependabot in #490
• Enable Embedding API in watsonx.ai by @andreadimaio in #508

Full Changelog: 0.12.0...0.12.1

0.12.0

What's Changed

• Mask the value of Set-Cookie headers by @jmartisk in #471
• Bump org.mvnpm.at.mvnpm:vaadin-webcomponents from 24.3.10 to 24.3.11 by @dependabot in #474
• Bump org.wiremock:wiremock-standalone from 3.5.2 to 3.5.3 by @dependabot in #470
• Bump org.mvnpm:lit from 3.1.2 to 3.1.3 by @dependabot in #469
• Improve testing coverage of multi-model cases by @geoand in #477
• Remove embedding model build item by @jmartisk in #480
• Added responseFormat to Azure OpenAi Chat Model by @UnvirtualHH in #483
• Allow user message to be a template by @geoand in #484
• Improve the csv-chatbot sample by @geoand in #485
• Fix the documentation about the memory of @RegisterAiService by @geoand in #486
• Add support for Azure AD Tokens by @csotiriou in #467
• Add qualifier to InProcessEmbeddingStore when needed 2 by @humcqc in #488
• Fix duplicate method exception when registering ai services by @iocanel in #492
• Add integration module with quarkus-websockets-next by @geoand in #487
• disable antora until having proper working solution by @maxandersen in #496

New Contributors

• @UnvirtualHH made their first contribution in #483
• @csotiriou made their first contribution in #467

Full Changelog: 0.11.1...0.12.0