- Adjust multi-modal memory write behavior for text and visual data
- Mask API keys in model list response to prevent exposure
- Add capability-based filtering to the model list API
- Add end_user_controller with unauthenticated endpoint for creating end users
- Implement get_or_create_end_user logic to handle duplicate end users by other_id
- Register end_user_controller router in main controller initialization
- Add list_memory_configs endpoint to retrieve all workspace memory configurations
- Update MemoryWriteRequest and MemoryReadRequest to make config_id required field
- Refactor memory API endpoints to parse request body directly instead of using Body parameter
- Add CreateEndUserRequest and CreateEndUserResponse schemas for end user creation
- Add ListConfigsResponse schema for configs listing endpoint
- Remove unused config_id and llm_model_id parameters from Neo4j write operation
- Update .gitignore to exclude redbear-mem-metrics and pitch-deck directories
1. Historical multimodal message writing is incorporated into the conversation context;
2. Resolve the issues where csv, json, and txt files cannot be recognized due to encoding problems;
3. File quantity limit;
4. Error details
1.Voice output is generated in a streaming manner.
2.Multimodal file storage type repair;
3.Adding features to the configuration of the sub-agents in the multi-agent system
* release/v0.2.6:
fix(web): ontology class default tag bugfix
fix(version): Version 0.2.6 Release Notes
fix(web): chat file delete bugfix
feat: support model load balancing and add message_id to API responses
feat: support model load balancing and add message_id to API responses
[changes] Work space isolation
[add] Recently, memory activities have adopted Redis caching.
[changes] Work space isolation
[add] Recently, memory activities have adopted Redis caching.
fix(web): upload add loading
[changes] The enumeration check has been changed to a string.
[changes] The enumeration check has been changed to a string.
feat(web): http-request add headers variable
fix(workflow): ensure file messages are written to messages in non-stream mode
fix(workflow): fix Dify compatibility issues
[changes] Memory write completion active failure interest cache
feat(workflow): support multimodal context
[changes] AI review and correction of code
[add] Semantic pruning is unified with the ontology engineering scenario.
feat(chat): add message_id field to chat API response
1. From the model square to the model list, the added models are initially set to be inactive. When manually activating them, a mandatory API key configuration is required.
2. Copying of applications (agent, workflow, multi_agent)
1. Increase support for visual models and multimodal models;
2. The application and workflow can input various multimodal files such as images, documents, audio, and videos.
