infiniflow/ragflow v0.23.0

on GitHub

v0.23.0

Released on December 27, 2025.

New features

• Memory
  • Implements a Memory interface for managing memory.
  • Supports configuring context via the Retrieval or Message component.
• Agent
  • Improves the Agent component's performance by refactoring the underlying architecture.
  • The Agent component can now output structured data for use in downstream components.
  • Supports using webhook to trigger agent execution.
  • Supports voice input/output.
  • Supports configuring multiple Retrieval components per Agent component.
• Ingestion pipeline
  • Supports extracting table of contents in the Transformer component to improve long-context RAG performance.
• Dataset
  • Supports configuring context window for images and tables.
  • Introduces parent-child chunking strategy.
  • Supports auto-generation of metadata during file parsing.
• Chat: Supports voice input.

Improvements

• Bumps RAGFlow's document engine, [Infinity](GitHub - infiniflow/infinity: The AI-native database built for LLM applications, providing incredibl) to v0.6.15 (backward compatible).

Data sources

• Google Cloud Storage
• Gmail
• Dropbox
• WebDAV
• Airtable

Model support

• GPT-5.2
• GPT-5.2 Pro
• GPT-5.1
• GPT-5.1 Instant
• Claude Opus 4.5
• MiniMax M2
• GLM-4.7.
• A MinerU configuration interface.
• AI Badgr (model provider).

API changes

HTTP API

• Converse with Agent returns complete execution trace logs.
• Create chat completion supports metadata-based filtering.
• Converse with chat assistant supports metadata-based filtering.

What's Changed

• fix cohere rerank base_url default by @HeyPhiS in #11353
• Feat: add OceanBase doc engine by @whhe in #11228
• fix(llm): handle None response in total_token_count_from_response by @zhangzhefang-github in #10941
• Feat: Add TCADP parser for PPTX and spreadsheet document types. by @aidansu in #11041
• Locale: update russian language by @rjohny55 in #11393
• locale en add russian language option by @rjohny55 in #11392
• Feature (canvas): Add mind tagging support by @huang-aoqin in #11359
• Use array syntax for commands in docker-compose-base.yml by @panzi in #11391
• Bump infinity to 0.6.6 by @yuzhichang in #11399
• Fix: Introducing a new JSON editor by @dcc123456 in #11401
• Fix:RagFlow not starting with Postgres DB by @buua436 in #11398
• Fix: Optimize edge check & incorrect parameter usage by @Magicbook1108 in #11396
• Feat: support union logic operations for meta data filters. by @KevinHuSh in #11404
• Fix: Change package-lock.json by @dcc123456 in #11407
• Fix: incorrect retrieval total count with pagination enabled by @yongtenglei in #11400
• Feat: Outputs data is directly synchronized to the canvas without going through the form. #10427 by @cike8899 in #11406
• Fix:improve multi-column document detection by @buua436 in #11415
• Feat: add more chunking method by @Magicbook1108 in #11413
• Fix: no result if metadata returns none. by @KevinHuSh in #11412
• Feat: enriches Notion connector by @yongtenglei in #11414
• Fix: missing parameters in by_plaintext method for PDF naive mode by @it-coder in #11408
• Fix: Modify the style of the user center #10703 by @dcc123456 in #11419
• component horizontal padding adjustment by @JimmyBenKlieve in #11418
• Feat: Add a loop variable to the loop operator. #10427 by @cike8899 in #11423
• Fix: UI adjustments, replacing private components with public components by @dcc123456 in #11438
• Fix: improve PDF text type detection by expanding regex content by @FallingSnowFlake in #11432
• Feat: ignore chunk size when using custom delimiters by @yongtenglei in #11434
• Feat: API supports toc_enhance. by @KevinHuSh in #11437
• Feat: Enable logical operators in metadata. #11387 #11376 by @cike8899 in #11442
• Fix:Resolves the issue of sessions not being saved when the variable is array. by @dcc123456 in #11446
• Feat: add auth header for Ollama chat model by @yongtenglei in #11452
• Add Moodle data source integration by @levischd in #11325
• Bump beartype to 0.22.6 by @yuzhichang in #11463
• Update logging for auto-generated SECRET_KEY by @FoolFu in #11458
• Fix nginx startup failure in HTTPS mode (host not found) by @davidlcarrascal in #11455
• Feat: Implement temporary conversation removal logic in ConversationD… by @Zwanan-github in #11454
• Fix:Modify the personal center style #10703 by @dcc123456 in #11470
• Fix: coroutine object has no attribute get by @Magicbook1108 in #11472
• Docs: Updated use_kg and toc_enhance switch descriptions by @writinwaters in #11485
• Allow create super user when start rag server. by @Ox0400 in #10634
• Feat: add datasource Dropbox by @yongtenglei in #11488
• Refactor: Email parser use with to handle buffer by @Woody-Hu in #11496
• Fix: After saving the model parameters of the chat page, the parameter disappears. #11500 by @cike8899 in #11501
• Fix: unexpected parameter. by @KevinHuSh in #11497
• Fix: Modify the style of your personal center #10703 by @dcc123456 in #11487
• Feat: support operator in/not in for metadata filter. by @KevinHuSh in #11503
• Feat: support operator in/not in for metadata filter. #11376 #11378 by @cike8899 in #11506
• Chore(deps): Bump pypdf from 6.0.0 to 6.4.0 by @dependabot[bot] in #11505
• Fix: code exec component vulnerability and add support for nested list and dict object by @yongtenglei in #11504
• Fix: uv lock updates by @KevinHuSh in #11511
• Feat: add addressing style config for S3-compatible storage by @jacy1101 in #11510
• Refactoring: Integrating the file preview component by @dcc123456 in #11523
• Fix: incorrect image merging for naive markdown parser by @yongtenglei in #11520
• Fix: excel default optimization. by @KevinHuSh in #11519
• Bump infinity to 0.6.7 by @yuzhichang in #11528
• Use Infinity single-field-multi-index by @yuzhichang in #11444
• Feat: Add Webdav storage as data source by @Jonah879 in #11422
• Fix: Refactoring and enhancing the functionality of the delete confirmation dialog component #10703 by @dcc123456 in #11542
• feat: improve metadata handling in connector service by @levischd in #11421
• Fix issue where HTML file parsing may lose content. by @myoldcat in #11536
• fix(files): replace hard coded status codes with constants by @huang-aoqin in #11544
• Minor style changes by @writinwaters in #11554
• Feat: add context for figure and table by @yongtenglei in #11547
• Fix ft_title_rag_fine by @yuzhichang in #11555
• Feat:new component Loop by @buua436 in #11447
• Feat: Add loop operator node. #10427 by @cike8899 in #11449
• Fix: enable structured output for agent with tool by @buua436 in #11558
• Feat: Delete useless knowledge base, chat, and search files. #10427 by @cike8899 in #11568
• Fix: remove garbage filtering rules by @buua436 in #11567
• Feat: add GPT-5.1, GPT‑5.1 Instant and Claude-Opus-4.5 by @yongtenglei in #11559
• update new models for tokenpony by @zhileihuang in #11571
• Fix: doc_aggs not correctly returned when no chunks retrieved. by @Magicbook1108 in #11578
• Feat: add gmail connector by @Magicbook1108 in #11549
• Feat: optimize meta filter generation for better structure handling by @TeslaZY in #11586
• Fix: Added styles for empty states on the page. #10703 by @dcc123456 in #11588
• [issue-11572]fix:metadata_condition filtering failed by @darion-yaphet in #11573
• Feat: Replace antd in the chat message with shadcn. #10427 by @cike8899 in #11590
• Fix: Error 102 "Can't find dialog by ID" when embedding agent with from=agent** #11552 by @cike8899 in #11594
• Feat: add child parent chunking method in backend. by @KevinHuSh in #11598
• Feat: create datasets from http api supports ingestion pipeline by @Magicbook1108 in #11597
• Feature: embedded chat theme by @Oranggge in #11581
• Fix typos by @omahs in #11607
• Docs: deploying a local model using Jina not supported by @writinwaters in #11624
• feat: add Redis username support by @coolbeevip in #11608
• fix: increase Quart RESPONSE_TIMEOUT and BODY_TIMEOUT for slow LLM responses by @dzikus in #11612
• Fix: [MinerU] Missing output file by @yongtenglei in #11623
• Fix: Table parse method issue. by @KevinHuSh in #11627
• Feat: support parent-child in search procedure. by @KevinHuSh in #11629
• Refa: make RAGFlow more asynchronous by @yongtenglei in #11601
• Fix: jina embedding issue by @Magicbook1108 in #11628
• Feat: Files uploaded via the dialog box can be uploaded without binding to a dataset. #9590 by @cike8899 in #11630
• Feat: support uploading in dialog. by @KevinHuSh in #11634
• Refactor: better describe how to get prefix for sync data source by @Woody-Hu in #11636
• Feat:new api /sequence2txt and update QWenSeq2txt by @buua436 in #11643
• Fix: file manager KB link issue. by @KevinHuSh in #11648
• Feat: Remove unnecessary dialogue-related code. #10427 by @cike8899 in #11652
• Refa: add MiniMax-M2 and remove deprecated MiniMax models by @yongtenglei in #11642
• Update upgrade guide: add stop server step and rename section by @yuzhichang in #11654
• Import rag_tokenizer from Infinity by @qinling0210 in #11647
• Feat: Delete useless request hooks. #10427 by @cike8899 in #11659
• Feat: add mineru auto installer by @Magicbook1108 in #11649
• feat: improve presentation PdfParser by @rommy2017 in #11639
• Refa: make RAGFlow more asynchronous 2 by @yongtenglei in #11664
• feat: improve Moodle connector functionality by @levischd in #11665
• Feature：Add voice dialogue functionality to the agent application by @dcc123456 in #11668
• Revert "Refa: make RAGFlow more asynchronous 2" by @KevinHuSh in #11669
• Docs: How to specify an ingestion pipeline when creating a dataset by @writinwaters in #11670
• Fix: incorrect async chat streamly output by @yongtenglei in #11679
• Feat:support tts in agent by @buua436 in #11675
• Feat: optimize the information displayed when .doc preview is unavailable by @Magicbook1108 in #11684
• Feat: support TOC transformer. by @KevinHuSh in #11685
• Refa: make RAGFlow more asynchronous 2 by @yongtenglei in #11689
• Feat: Replace antd with shadcn and delete the template node. #10427 by @cike8899 in #11693
• Refactor code by @JinHai-CN in #11694
• Feat: RAG evaluation by @hsparks-codes in #11674
• Standardize UI text capitalization to sentence case by @redredrrred in #11696
• feat: Auto-disable Raptor for structured data (Issue #11653) by @hsparks-codes in #11676
• Fix: handle MinerU sanitized filenames when reading output by @yongtenglei in #11701
• Update since 'File manager' is renamed to 'File' by @JinHai-CN in #11698
• Feat: confluence space key by @hsparks-codes in #11706
• Fix: Correct pagination and early termination bugs in chunk_list() by @dnosferatus in #11692
• Refactor: Remove useless for and add by @Woody-Hu in #11720
• feat(gcs): Add support for Google Cloud Storage (GCS) integration by @dafa-wiratama in #11718
• Fix: relative page_number in boxes by @rommy2017 in #11712
• fix: resolve "'bool' object has no attribute 'items'" in SDK enabled … by @shirukai in #11725
• Feature：Add a loading status to the agent canvas page. by @dcc123456 in #11733
• Feat: detect docx support via header-byte inspection by @Magicbook1108 in #11731
• Fix: none type error. by @KevinHuSh in #11735
• Refa: cleanup synchronous functions in agent_with_tools by @yongtenglei in #11736
• Remove huqie.txt from RAGFflow and bump infinity to 0.6.10 by @qinling0210 in #11661
• Change docker container restart policy by @JinHai-CN in #11695
• Doc: Updated Create dataset descriptions by @writinwaters in #11742
• Feat: add more attribute for confluence connector. by @Magicbook1108 in #11743
• Feat: update front end for confluence connector by @Magicbook1108 in #11747
• Delete:remove unused tests by @buua436 in #11749
• Feat: Display the ID of the code image in the dialog. #10427 by @cike8899 in #11746
• fix(sdk): add cancel_all_task_of call in stop_parsing endpoint by @tedhappy in #11748
• Feat: use filepath for files with the same name by @Jonah879 in #11752
• Add ut test into CI by @JinHai-CN in #11753
• Feat: Users can chat directly without first creating a conversation. #11768 by @cike8899 in #11769
• Fix: Newly added models to OpenAI-API-Compatible are not displayed in the LLM dropdown menu in a timely manner. #11774 by @cike8899 in #11775
• fix(mcp-server): Ensure all document meta-data is cached by @gileslloyd in #11767
• Fix MinerU API sanitized-output lookup and manual chunk tuple handling by @shaoqing404 in #11702
• Feat: Fillup component return value not object by @TeslaZY in #11780
• Feature (canvas): Add Api for download "message" component output's file by @huang-aoqin in #11772
• Fix errors by @JinHai-CN in #11795
• Fix:toc in pipeline by @buua436 in #11785
• Refa: cleanup synchronous functions in chat_model and implement synchronization for conversation and dialog chats by @yongtenglei in #11779
• Add huggingface-hub dependency by @mstfldmr in #11794
• Features: Memory page rendering and other bug fixes by @dcc123456 in #11784
• fix: Correct toast type import path in use-toast hook by @Rohit10jr in #11791
• Refactor:book parser use with to handle bytesIO by @Woody-Hu in #11800
• Feat: default start admin by @Lynn-Inf in #11801
• Fix: Changed 'HightLightMarkdown' to 'HighLightMarkdown' by @dcc123456 in #11803
• Fix errors by @JinHai-CN in #11804
• Fix: chunk method error. by @KevinHuSh in #11807
• Fix: advanced_ingestion_pipeline by @ltgkb in #11816
• Fix: parent-child chunking method by @buua436 in #11810
• fix: align http client proxy kwarg by @levischd in #11818
• fix: set default embedding model for TEI profile in Docker deployment by @SJpoor in #11824
• Fix:[ERROR][Exception]: list index out of range by @Woody-Hu in #11826
• Refa: migrate CV model chat to Async by @yongtenglei in #11828
• Feature: Memory interface integration testing by @dcc123456 in #11833
• Bump infinity to v0.6.11. Requires python>=3.11 by @yuzhichang in #11814
• fix: prevent redundant retries in async_chat_streamly upon success by @N0bodycan in #11832
• Docs: Enhance API reference for file management by @changkeke in #11827
• feat: add Italian language translation support by @PentaFrame-Development in #11844
• Fix: The variables in the message node are not displaying correctly. #11839 by @cike8899 in #11841
• Refa: treat MinerU as an OCR model by @yongtenglei in #11849
• Fix: duplicate output by async_chat_streamly by @Magicbook1108 in #11842
• Refa:replace trio with asyncio by @buua436 in #11831
• Bump python to >=3.12 by @yuzhichang in #11846
• executor_manager updated docker version by @bluthen in #11806
• Feat/memory by @Lynn-Inf in #11812
• Fix:Modify the name of the Overlapped percent field by @dcc123456 in #11866
• Fix:csv parse in Table by @buua436 in #11870
• Refa:remove sensitive information by @buua436 in #11873
• feat: enhance OBConnection.search by @whhe in #11876
• Feat: Add configuration for webhook to the begin node. #10427 by @cike8899 in #11875
• Fix RuntimeError: asyncio.run() cannot be called from a running event loop when calling mindmap endpoint. by @N0bodycan in #11880
• Feat: enhance metadata operation by @yongtenglei in #11874
• MinerU supports for the new backend vlm-mlx-engine by @TeslaZY in #11864
• Fix data_sync startup crash by properly invoking async main by @davidlcarrascal in #11879
• Added semi-automatic mode to the metadata filter by @TeslaZY in #11886
• Fix:async issue and sensitive logging by @buua436 in #11895
• Add DeepseekV3.2 of Tongyi-Qianwen and remove unused code by @TeslaZY in #11898
• Refa: treat MinerU as an OCR model 2 by @yongtenglei in #11905
• Feat: Add mineru as a model manufacturer to the system. #10621 by @cike8899 in #11903
• Fix: tokenizer issue. by @KevinHuSh in #11902
• feat: Add Single Bucket Mode for MinIO/S3 by @arogan178 in #11416
• Feat: Flatten the request schema of the webhook #10427 by @cike8899 in #11917
• Feat: Add box connector by @Magicbook1108 in #11845
• Docs: How to use restful API to update or delete metadata by @writinwaters in #11912
• Fix: correct metadata update behavior by @yongtenglei in #11919
• Feature/docs generator by @PentaFrame-Development in #11858
• Fix: forget-reset password by @Magicbook1108 in #11927
• Feat: Displaying the file option in the webhook's request body #10427 by @cike8899 in #11928
• Refa: refactor metadata filter by @yongtenglei in #11907
• Feat: Add GPT-5.2 & pro by @Magicbook1108 in #11929
• Fix: raptor don't have attribute chat by @yongtenglei in #11936
• Fix: aspose-slide issue. by @KevinHuSh in #11935
• Fix: Asure-OpenAI resource not found by @yongtenglei in #11934
• Fix: presentation parsing & Embedding encode exception handling by @Magicbook1108 in #11933
• Support uploading encrypted files to object storage (#11837) by @yngvarhuang in #11838
• Feat: Set the return value of the webhook to a string. #10427 by @cike8899 in #11945
• Doc: executor manager updated docker version by @yongtenglei in #11946
• Refactor: Improve the logic to calculate embedding total token count by @Woody-Hu in #11943
• Fix: pipeline ignore MinerU backend config and vllm module is missing by @yongtenglei in #11955
• feat: display name instead of key in user fillup form submission by @zhulenghan in #11931
• Update dataset config and retrieval testing by @JinHai-CN in #11958
• Fix: Bug fixes by @dcc123456 in #11960
• Fix: 'AzureEmbed' object has no attribute 'total_token_count_from_response' by @Woody-Hu in #11962
• Remove unused py module dependencies by @JinHai-CN in #11964
• Feature/mineru improvements by @concertdictate in #11938
• Feature/excel export fix by @Shivamjohri247 in #11914
• Fix security issue by @JinHai-CN in #11965
• Docs: Migrate to single bucket mode (contributed by community) by @writinwaters in #11972
• Feat: Enable image edit in edit_chunk by @Magicbook1108 in #11971
• Fix error and format issue by @JinHai-CN in #11975
• Docs: Updated executor manager prerequisites by @writinwaters in #11978
• Change knowledge base to dataset by @JinHai-CN in #11976
• feat: add ingestion pipeline children delimiters configs by @JimmyBenKlieve in #11979
• Feature：memory function complete by @dcc123456 in #11982
• Update default prompt by @JinHai-CN in #11984
• Refa: only support MinerU-API now by @yongtenglei in #11977
• fix: unable to save ingestion pipeline config without modifying children delimiter by @JimmyBenKlieve in #11991
• Fix: update document api sample reponse is out of date. by @Magicbook1108 in #11989
• Fix: Fixed the issue of empty memory parameters by @dcc123456 in #11988
• Ref: update loggings by @Magicbook1108 in #11987
• Fix: add multimodel models in chat api by @Magicbook1108 in #11986
• feat: add toc option in transformer node in ingestion pipeline by @JimmyBenKlieve in #11992
• Feat: Reject default admin account log in to normal services by @Magicbook1108 in #11994
• Fix: parent-children chunking method. by @KevinHuSh in #11997
• Add license and Fix IDE warnings by @JinHai-CN in #11985
• Fix table format warning in Markdown file by @JinHai-CN in #12002
• Fix: duplicated PDF parser by @yongtenglei in #12000
• Fix: model not authorized by @yongtenglei in #12001
• feat: add image uploader in edit chunk dialog by @JimmyBenKlieve in #12003
• Refactor: Improve the timestamp consistency by @Woody-Hu in #11942
• task executor issues by @concertdictate in #12006
• Refactor: improve cohere calculate total counts by @Woody-Hu in #12007
• Fix IDE warnings by @JinHai-CN in #12010
• Fix: image edit in edit_chunk by @Magicbook1108 in #12009
• Refact: Update loggings by @Magicbook1108 in #12014
• Feat: trace information can be returned by the agent completion API by @yongtenglei in #12019
• Feat: bedrock iam authentication by @Magicbook1108 in #12020
• Feat: use filepath for files with the same name for all data source types by @Jonah879 in #11819
• Feat:update webhook component by @buua436 in #11739
• Feat: Further update Bedrock model configs by @Magicbook1108 in #12029
• Refa: better UX for adding OCR model by @yongtenglei in #12034
• Feat: chats completions API supports metadata filtering by @yongtenglei in #12023
• Feat: Display error messages from intermediate nodes of the webhook. #10427 by @cike8899 in #11954
• Docs: How to call MinerU as a remote service by @writinwaters in #12004
• Feat: Display error messages from intermediate nodes. #10427 by @cike8899 in #12038
• Add AI Badgr as OpenAI-compatible chat model provider by @miguelmanlyx in #12018
• Feat: Images appearing consecutively in the dialogue are merged and displayed in a carousel. #10427 by @cike8899 in #12051
• Feature: Implement metadata functionality by @dcc123456 in #12049
• Include document_id in knowledgebase info retrieval by @haowenj in #12041
• Feat: document list and filter supports metadata filtering by @yongtenglei in #12053
• feat: supports multiple retrieval tool under an agent by @JimmyBenKlieve in #12046
• Feat: Separate connectors from s3 by @Magicbook1108 in #12045
• feat: add optional cache busting for image by @JimmyBenKlieve in #12055
• Feat: enhance Excel image extraction with vision-based descriptions by @buua436 in #12054
• Fix: Python SDK retrieve document_name is empty by @Woody-Hu in #12062
• Fix: update RAGFlow SDK for consistency by @Magicbook1108 in #12065
• Remove Chinese comments and fix function arguments errors by @JinHai-CN in #12052
• Fix: list MCP tools may block by @yongtenglei in #12067
• Refactor:remove useless try catch for ppt parser by @Woody-Hu in #12063
• Potential fix for code scanning alert no. 57: Clear-text logging of sensitive information by @yingfeng in #12071
• Potential fix for code scanning alert no. 58: Clear-text logging of sensitive information by @yingfeng in #12070
• Potential fix for code scanning alert no. 60: Clear-text logging of sensitive information by @yingfeng in #12068
• Fix character escape by @JinHai-CN in #12072
• Potential fix for code scanning alert no. 59: Clear-text logging of sensitive information by @yingfeng in #12069
• Feat: When the webhook returns a field in streaming format, the message displays the status field. #10427 by @cike8899 in #12075
• Images that appear consecutively in the dialogue are displayed using a carousel. #12076 by @cike8899 in #12077
• fix: guard Dashscope response attribute access in token/log utils by @yongtenglei in #12082
• feat: display chunk type in chunk editor and dialog by @JimmyBenKlieve in #12086
• Fix IDE warnings by @JinHai-CN in #12085
• Feature/1217 by @dcc123456 in #12087
• fix: hide drop-zone upload button when picked an image by @JimmyBenKlieve in #12088
• Feat: optimize aws s3 connector by @Magicbook1108 in #12078
• Refactor add_llm and add speech to text by @JinHai-CN in #12089
• feat: add image context window in dataset config by @JimmyBenKlieve in #12094
• Feat: output intinity test log by @Lynn-Inf in #12097
• Fix: Tika server upgrades. by @KevinHuSh in #12073
• Feat: enhance webhook response to include status and success fields and simplify ReAct agent by @buua436 in #12091
• Fix: fix task cancel by @Magicbook1108 in #12093
• Update workflow by @JinHai-CN in #12101
• Feat: Hide part of the message field in webhook mode #10427 by @cike8899 in #12100
• Fix:Bugs fix (Reduce metadata saving steps ...) by @dcc123456 in #12095
• Fix: vision_figure_parser_docx/pdf_wrapper by @Magicbook1108 in #12104
• Feat: Add memory multi-select dropdown to recall and message operator forms. #4213 by @cike8899 in #12106
• Feat: add GLM-4.7 by @yongtenglei in #12102
• Fix: update method call to use simplified async tool reaction by @buua436 in #12108
• Fix only one of multiple retrieval tools is effective by @TeslaZY in #12110
• Fix: Metadata bugs. by @dcc123456 in #12111
• Feat: Translate the text of the webhook debugging interface. #10427 by @cike8899 in #12115
• redirect stderr to stdout by @JinHai-CN in #12122
• Fix: metadata issue & graphrag speeding up. by @KevinHuSh in #12113
• Refactor: enhance next_step prompt by @buua436 in #12117
• fix: transformer toc prompt text incorrect by @JimmyBenKlieve in #12116
• fix: add null safety checks in webhook response status hook by @SalmonWu13 in #12114
• refactor: reword locale translations by @JimmyBenKlieve in #12118
• Fix: The form waiting for input is not displayed in the dialog message. #12129 by @cike8899 in #12130
• Refa: image table context window by @yongtenglei in #12132
• Fix: /kb/update does not update FileService by @Magicbook1108 in #12121
• Fix test error by @JinHai-CN in #12194
• Fix: toc no chunk found issue. by @KevinHuSh in #12197
• Feat: supports filter documents by empty metadata by @yongtenglei in #12180
• Fix directory name by @JinHai-CN in #12195
• Fix: keep behavior consistent for converse_with_chat_assistant by @yongtenglei in #12190
• Refa: remove MinerU settings from .env by @yongtenglei in #12201
• Update german language file (resubmission) by @hauberj in #12208
• Feat: Images referenced in chat messages are displayed as a carousel at the bottom of the message. #12076 by @cike8899 in #12207
• Fix: Metadata tips info by @dcc123456 in #12209
• Fix 404 not found error in CI by @JinHai-CN in #12200
• fix: chunk editor allows update image only if chunk type is image by @JimmyBenKlieve in #12210
• Feat: add Airtable connector and integration for data synchronization by @buua436 in #12211
• Fix: gen metadata error. by @KevinHuSh in #12212
• Fix: Api key modal bug by @dcc123456 in #12213
• Feat: An image carousel is displayed at the bottom of the agent's chat messages. #12076 by @cike8899 in #12215
• Feat: Hide the autoplay switch for message operators in webhook mode. #10427 by @cike8899 in #12216
• Feat: message manage by @Lynn-Inf in #12196
• Fix: Can not select LLM in memory page by @dcc123456 in #12219
• Fix text issue by @JinHai-CN in #12221
• Feat: The agent can only retrieve content from the knowledge base or memory. #4213 by @cike8899 in #12224
• Feat: Preview the image at the bottom of the message #12076 by @cike8899 in #12225
• Fix: Memory-related bug fixes by @dcc123456 in #12226
• Fix: judge retrieval from by @Lynn-Inf in #12223
• Fix forget policy by @JinHai-CN in #12228
• Fix memory config: user prompt text box by @JinHai-CN in #12229
• Fix small issues on UI by @JinHai-CN in #12231
• Fix message duration by @JinHai-CN in #12233
• Fix: memory by @Lynn-Inf in #12230
• Update deploy_local_llm.mdx with vllm guide support by @yiminghub2024 in #12222
• Feat: Display the selected list of memories in the retrieval node. #4213 by @cike8899 in #12235
• Fix: allow update memory type by @Lynn-Inf in #12237
• Fix: Memory-related bug fixes by @dcc123456 in #12238
• Fix: rm field not allow check by @Lynn-Inf in #12240
• Refa: improve image table context by @yongtenglei in #12244
• Fix: Memory sava issue by @dcc123456 in #12243
• Fix: parent-children pipleine bad case. by @KevinHuSh in #12246
• Docs: v0.23.0 release notes by @writinwaters in #12251
• Fix: order by when pick msg to rm by @Lynn-Inf in #12247
• Fix: fifo -> FIFO by @dcc123456 in #12257
• Fix: frontend cannot sync document window context by @yongtenglei in #12256
• Fix: optimize init memory_size by @Lynn-Inf in #12254
• Docs: Fixed a display issue by @writinwaters in #12259
• Bump infinity to 0.6.15 by @JinHai-CN in #12264
• Fix memory issue on Infinity 0.6.15 by @JinHai-CN in #12258
• Docs: Update version references to v0.23.0 in READMEs and docs by @asiroliu in #12253

  New Contributors

  • @HeyPhiS made their first contribution in #11353
  • @zhangzhefang-github made their first contribution in #10941
  • @it-coder made their first contribution in #11408
  • @FallingSnowFlake made their first contribution in #11432
  • @levischd made their first contribution in #11325
  • @FoolFu made their first contribution in #11458
  • @davidlcarrascal made their first contribution in #11455
  • @Zwanan-github made their first contribution in #11454
  • @Ox0400 made their first contribution in #10634
  • @jacy1101 made their first contribution in #11510
  • @myoldcat made their first contribution in #11536
  • @darion-yaphet made their first contribution in #11573
  • @Oranggge made their first contribution in #11581
  • @omahs made their first contribution in #11607
  • @coolbeevip made their first contribution in #11608
  • @dzikus made their first contribution in #11612
  • @qinling0210 made their first contribution in #11647
  • @rommy2017 made their first contribution in #11639
  • @hsparks-codes made their first contribution in #11674
  • @dnosferatus made their first contribution in #11692
  • @dafa-wiratama made their first contribution in #11718
  • @shirukai made their first contribution in #11725
  • @tedhappy made their first contribution in #11748
  • @gileslloyd made their first contribution in #11767
  • @shaoqing404 made their first contribution in #11702
  • @mstfldmr made their first contribution in #11794
  • @Rohit10jr made their first contribution in #11791
  • @ltgkb made their first contribution in #11816
  • @SJpoor made their first contribution in #11824
  • @N0bodycan made their first contribution in #11832
  • @changkeke made their first contribution in #11827
  • @PentaFrame-Development made their first contribution in #11844
  • @zhulenghan made their first contribution in #11931
  • @concertdictate made their first contribution in #11938
  • @Shivamjohri247 made their first contribution in #11914
  • @miguelmanlyx made their first contribution in #12018
  • @haowenj made their first contribution in #12041
  • @SalmonWu13 made their first contribution in #12114
  • @hauberj made their first contribution in #12208

  Full Changelog: v0.22.1...v0.23.0