This commit introduces support for the Anthropic Claude3 Message API
(https://api.anthropic.com), enabling direct interaction with its services.
This is not a Bedrock Anthropic Claude3 implemenation.
Changes include:
- Implementation of a low-level client, AnthropicApi, to interact with
the message API endpoints specified in the Anthropic documentation
(https://docs.anthropic.com/claude/reference/messages_post), including support for streaming.
- Addition of AnthropicApi tests to ensure functionality and reliability.
- Support for multimodal requests within AnthropicApi.
- Adding the spring-ai-anthropic and boot starter into BOM and parent POM modules for streamlined usage.
- Add Anthropic Auto-configuration and Boot Starter for seamless integration into existing projects.
- Implementation of AnthropicChatClient with capabilities for synchronous and streaming communication,
including support for multimodal messages.
- Inclusion of both unit and integration tests to validate functionality across various scenarios.
- Add Antora documentation with comprehensive guidance on using AnthropicApi and AnthropicChatClient.
- Add of Ahead-of-Time (AOT) hints for AnthropicApi.
- update anthropic diagram
- Implemented Bedrock Jurassic ChatClient
- added documentation reference
- implement auto-configuration and boot starter
- Disable the BedrockAi21Jurassic2ChatClientIT.emojiPenaltyWhenTrueByDefaultApplyPenaltyTest() test
as it fails when run in combination with the other tests.
- tename /api/clients/ into /api/chat
- move the the image from /api/clients to /api
- fix the layout inside the chat and embeddings docs. Moving the runtime options and sample controllers at top level.
- adjust all affected links.
* Fix Typo: Duplicate 'for' in documentation text
* Fix Typo: Duplicate 'to' in documentation text
* Fix broken links in documentation
* Correct grammar by deleting unnecessary 'an' in documentation
* Fix typo: Change 'tunning' to 'tuning' in documentation
* Fix typo: Change 'an' to 'can' in documentation
* Fix typo: Change 'generats' to 'generates' in documentation
* Fix grammatical error: Change 'a AI' to 'an AI' in documentation
* Fix grammatical error: Change 'a AI' to 'an AI' in code
* Fix Typo: Duplicate 'for' in code
* Azure uses 'deployment-name' when provisioning models and is what needs to
be passed in to the client, not the model name.
This is a difference with the OpenAI API that doesn't have a deployment-name
This change aligns the terminology used with Azure so that there is
less confusion when setting configuration property values
Fixes#10
- Establish a new "spring-ai-retry" project, implementing a default HTTP error handler,
RetryTemplate, and handling both Transient and Non-Transient Exceptions.
- Streamline existing clients (e.g., OpenAI and MistralAI) to utilize "spring-ai-retry."
- Integrate retry auto-configuration with customizable properties, extending it to OpenAI and MistralAI Auto-Configs.
- Allow configuration of RetryTemplate and ResponseErrorHandler for various clients, including OpenAIChatClient,
OpenAiEmbeddingClient, OpenAiAudioTranscriptionCline, OpenAiImageClient, MistralAiChatClient, and MistralAiEmbeddingClient.
- Add tests for default RestTemplate and ResponseErrorHandler configurations in OpenAI and MistralAI.
- Introduce new retry auto-config properties: "onClientErrors" and "onHttpCodes".
- Implement tests for retry auto-config properties.
- Generate missing license headers.
Also
- fix a MistralAiApi (and OpenAiApi) issue just introduced by Mistral API
- change the auto-config properties prefix from spring.ai.mistral.* to spring.ai.mistralai.*
- minor doc references updates.
- Implement QdrantVectorStore.
Uses a custom parser for converting Spring AI metadata(Map<String, Object>) to Qdrant GRPC payload.
- Implement Qdrant Expression Filter support.
Uses a custom parser for converting Spring AI filters to Qdrant-compatible GRPC filters.
- Add ITs using testcontainers.
- Add antora docs adrant.adoc.
- Add Qdrant vector store auto-configuraton and boot starter.
Additional (review) change:
- Fix poms parent to 0.8.1-SNAPSHOT.
- Rename ObjectFactory into QdrantObjectFactor.
- Rename ValueFactory into QdrantValueFactory.
- Move the org.springframework.ai.vectorstore package into org.springframework.ai.vectorstore.qdrant.
- Add missing Autoconfigure definition.
- Add missing license and JavaDocs.
- Minor code style improvmentes.
- Move the qdrant version to the main pom
- Add QdrantVectorStoreAutoConfigurationIT
- Remove guava dependency
- Improve gdrant.adoc conent and structure.
- Remove the grpc-protobuf dependency
Resolves#331
- Extend the Spring AI Message with getMediaData() : List<MediaData>
MediaData is a pair of MimeType and data of type Object.
Message#getContent() return text only.
- VertexAI Gemini Support
- implement VertexAiGeiminChatClient for ChatClient and StreamingChat client and support for MediaData content.
add IT tests for Chat, Streaming and Multimodality
- add Auto-configuration + ITs
- add Gemini Spring Boot starter.
- add clients and boot starters to the Spring AI BOM.
- add Anotra documentation for the Gemini chat client.
- update gemini to latest 26.33.0 BOM.
- add vertex ai gemini dependencies to the BOM.
- add Vertex AI Gemini API Function Calling support
- add Gemini API Function Calling Streaming support
- add vertex ai gemini function calling documentation
- factor out the Function Calling functionality into common abstraction used by OpenAI, Azure and Gemini.
- group the VertexAI documentation under a common parent
- add PortableFunctionCallingOption that implements FunctionCallingOptions and ChatOptions and provide builder for it.
- remove some deprecated code.
- allow authorization with GoogleCredentials form json file.
- add AOT support for VertexAI Gemini.
- move legacy Vertex AI into VertexAI PaLM2.
- better handling for empty chat responses.
- update the Gemini version to latest 26.33.0. This required lifting the protobuf-java to 3.25.2 as well.
- fix a bug for handling System messages with Gemini.
- Implement Azure OpenAI Function Calling
Uses the same the common abstractions used by OpenAI and Gemini: FunctionCallingOptions and AbstractFunctionCallSupport
Property is of format `*.enabled=` and is true by default.
Note: the Bedrock clients are disabled by default.
- Add deactivation support for Vertex AI PaLM2 clients
- Add deactivation support for Azure OpenAI clients
- Add deactivation support for huggingface chat client
- Add deactivation support for Ollama clients
- Add deactivation support for OpenAI clients
- Add deactivation support for Stability Image and PostgresML clients
- Add deactivation support for ONNX Transformers Embedding client
Resolves#363
* Added instructions on how to set up and use the 'Deployment Name' for Azure AI applications in the documentation.
* Corrected the default chat model from 'gpt-3.5-turbo' to 'gpt-35-turbo' for alignment with Azure's naming conventions.
