49 Pull requests merged by 16 people

• Fix:Unimplemented function in ppt_parser

  #8095 merged Jun 6, 2025

• Docs: Updated server launching code

  #8093 merged Jun 6, 2025

• Test: add sdk Document test cases

  #8094 merged Jun 6, 2025

• Fix(python-sdk): Add name filtering support to Dataset.list_documents()

  #8090 merged Jun 5, 2025

• Fix: document typo in test

  #8091 merged Jun 5, 2025

• Fix: Resolve JSON download errors in Document.download()

  #8084 merged Jun 5, 2025

• Fix: Fixed an issue where using the new quote markers would cause dialogue output to have delete symbols #7623

  #8083 merged Jun 5, 2025

• Feat: Convert the inputs parameter of the begin operator #3221

  #8081 merged Jun 5, 2025

• Fix:when stream=false，new message without sessionid does no

  #8078 merged Jun 5, 2025

• Feat: Solved the problem that BeginForm would get stuck when modifying data #3221

  #8080 merged Jun 5, 2025

• Test: add sdk Dataset test cases

  #8077 merged Jun 5, 2025

• Fix: issue for tavily only in a assistant.

  #8076 merged Jun 5, 2025

• Fix: Document.update() now refreshes object data

  #8068 merged Jun 5, 2025

• Fix: set output for Message template

  #8064 merged Jun 5, 2025

• Fix: Authentication Bypass via predictable JWT secret and empty token validation

  #7998 merged Jun 5, 2025

• Update api_utils.py

  #8069 merged Jun 5, 2025

• Feat: Create empty agent #3221

  #8054 merged Jun 5, 2025

• Fix: remove zhipu ai api key

  #8066 merged Jun 5, 2025

• Don't use '，' (U+FF0C) but ', ' (U+2C U+20)

  #8063 merged Jun 5, 2025

• Fix: DataSet.update() now refreshes object data

  #8058 merged Jun 5, 2025

• Fix: data missing after upgrading.

  #8047 merged Jun 4, 2025

• Feat: Add RunSheet component #3221

  #8045 merged Jun 4, 2025

• Feat: Allow update conversation parameters and persist to database in completion

  #8039 merged Jun 4, 2025

• Fix: Allow None value for parser_config in create_dataset SDK method

  #8041 merged Jun 4, 2025

• Refa: Move HTTP API tests to top-level test directory

  #8042 merged Jun 4, 2025

• Fix: /v1/conversation/completion KeyError: 'conversation_id'

  #8037 merged Jun 4, 2025

• Docs: Miscellaneous UI updates

  #8031 merged Jun 4, 2025

• Fix: Grammar and clarity improvements in prompt templates

  #8023 merged Jun 3, 2025

• Feat: Add DynamicPrompt component #3221

  #8028 merged Jun 3, 2025

• Feat: Add AgentNode component #3221

  #8019 merged Jun 3, 2025

• Fix: no kb selected for an assistant.

  #8021 merged Jun 3, 2025

• Fix: Set Output In Category Component

  #8010 merged Jun 3, 2025

• Feat: Construct RetrievalForm with original fields #3221

  #8012 merged Jun 3, 2025

• Fix: sdk can not update chat model

  #8016 merged Jun 3, 2025

• Feat: sync test group to top pyproject.toml

  #8015 merged Jun 3, 2025

• Fix: url path join issue.

  #8013 merged Jun 3, 2025

• Feat: add voyage-multimodal-3

  #7987 merged Jun 3, 2025

• Update synonym dictionary file

  #7997 merged Jun 3, 2025

• Docs: Updated UI tips for reranker

  #7983 merged May 30, 2025

• Feat: Add the example component of the classification operator #3221

  #7986 merged May 30, 2025

• Fix: order chunks from docx by positions.

  #7979 merged May 30, 2025

• Feat: Use one-way data flow to synchronize the form data to the canvas #3221

  #7977 merged May 30, 2025

• Fix: Unnecessary truncation in markdown parser

  #7972 merged May 30, 2025

• Update upload filename length limit from 128 to 256, which is aligned with os

  #7971 merged May 30, 2025

• Fix: Prevent Flask hot reload from hanging due to early thread startup

  #7966 merged May 30, 2025

• Fix: Add user_id when create Conversation

  #7960 merged May 30, 2025

• Update docker-compose.yml

  #7962 merged May 30, 2025

• Fix Bug: When performing the dify_retrieval, the metadata of the document was empty.

  #7968 merged May 30, 2025

• fix: resolve residual image files issue after document deletion

  #7964 merged May 30, 2025

4 Pull requests opened by 4 people

• fix: Increase timeout time of mysql to avoid false unhealthy check

  #7975 opened May 30, 2025

• Feat: "api_4_conversation" adds the same 'name' column as the "conversation"

  #8046 opened Jun 4, 2025

• Fixed baidu request

  #8072 opened Jun 5, 2025

• Feat: Receive reply messages of different event types from the agent #3221

  #8100 opened Jun 6, 2025

18 Issues closed by 10 people

• [Bug]: Unimplemented function in ppt_parser

  #8088 closed Jun 6, 2025

• [Feature Request]: Allow dify to modify the vector_similarity_weight？

  #7767 closed Jun 6, 2025

• [Question]: Update chunk:Q&A must be separated by TAB/ENTER key

  #8075 closed Jun 5, 2025

• [Question]: In version 0.19, the Python API does not support calling custom embedded models.

  #8059 closed Jun 5, 2025

• [Bug]: Python SDK Document.update() doesn't refresh object's internal data

  #8067 closed Jun 5, 2025

• [Bug]: Python SDK DataSet.update() doesn't refresh object's internal data

  #8057 closed Jun 5, 2025

• [Bug]: Error when inviting a new user hint : 102 User not found.

  #4707 closed Jun 4, 2025

• [Question]: reference.chunks.positions numbers meaning

  #8056 closed Jun 4, 2025

• 调用/api/v1/retrieval API时怎样提供rerank_id参数

  #8017 closed Jun 4, 2025

• [Bug]: /v1/conversation/completion KeyError: 'conversation_id'

  #8033 closed Jun 4, 2025

• [Bug]:Unable to add the Volcano Engine embedding model

  #8025 closed Jun 3, 2025

• [Bug]: The RagFlow HTTP API for updating the chat assistant cannot update the model.

  #7791 closed Jun 3, 2025

• [Bug]: Using os.path.join for api_base and v1 causes incorrect URL on Windows

  #7980 closed Jun 3, 2025

• [Question]: ver:0.19 , api list chat assistant session error

  #7921 closed Jun 3, 2025

• [Question]: 请问支持在昇腾910B上部署吗？

  #7944 closed May 30, 2025

• [Question]: 请问增强的图像显示这个效果要怎样看到？

  #7967 closed May 30, 2025

• [Question]: The order of the chunk parsed by General is inconsistent with that of the original text

  #7943 closed May 30, 2025

• [Bug]: Code Component Can not send http request

  #7958 closed May 30, 2025

59 Issues opened by 48 people

• [Question]: ssl.SSLCertVerificationError: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate

  #8098 opened Jun 6, 2025

• [Bug]: QWenCV failed to extract text from structured streaming responses

  #8097 opened Jun 6, 2025

• [Bug]: BlockingIOError: [Errno 11] write could not complete without blocking

  #8096 opened Jun 6, 2025

• [Question]: Why not use minerU to parse the document? At least this option can be added?

  #8092 opened Jun 5, 2025

• [Question]: I want to use ragflow as an external RAG for knowledge retrieval. How should I integrate it into my existing AI framework?

  #8089 opened Jun 5, 2025

• [Question]: QA Chunk Embedding and File Management API

  #8087 opened Jun 5, 2025

• [Question]: Why is it not necessary to set up an embedding model when creating a chat assistant?

  #8086 opened Jun 5, 2025

• [Question]: Get real images from pdf

  #8085 opened Jun 5, 2025

• [Question]: When I select an img2txt model for PDF document parsing, what task should I choose when deploying this model?

  #8079 opened Jun 5, 2025

• [Feature Request]: Object storage supports OpenDAL.

  #8074 opened Jun 5, 2025

• [Feature Request]:optional ai model during chunk creation

  #8071 opened Jun 5, 2025

• [Bug]: when stream=false，new message without sessionid does no

  #8070 opened Jun 5, 2025

• [Bug]: graphrag task doesn't create

  #8065 opened Jun 5, 2025

• [Question]: Is this meant to be '，' (U+FF0C)?

  #8062 opened Jun 4, 2025

• [Feature Request]: user_default_llm supports to configure different model factory for each type model

  #8061 opened Jun 4, 2025

• [Bug]: Text excerpts from file prompts.py contianed in generated answer

  #8060 opened Jun 4, 2025

• [Bug]: delimiter not working when parse docx pdf and markdown

  #8055 opened Jun 4, 2025

• [Bug]: /api/v1/retrieval: dataset_id format (string vs. array) depends on chunk ingestion source (/chunk/create vs. file import)

  #8052 opened Jun 4, 2025

• [Bug]: Okta OIDC integration would return jwks 404

  #8051 opened Jun 4, 2025

• [Bug]: Knowledge base is selected, but not referenced in chat

  #8050 opened Jun 4, 2025

• [Question]: Code component read timeout

  #8049 opened Jun 4, 2025

• [Feature Request]: add AI/ML API as provider

  #8048 opened Jun 4, 2025

• [Question]: Incorrect RAGFlow Chunking: Headings and Tables Split

  #8044 opened Jun 4, 2025

• [Question]: official online demo - openai api style access - local certificate issues

  #8043 opened Jun 4, 2025

• [Question]: can not web login both with edge and chrome

  #8040 opened Jun 4, 2025

• [Feature Request]: add migrate tool for "doc engine from Elasticsearch to Infinity"

  #8038 opened Jun 4, 2025

• [Feature Request]: Return number of used tokens from endpoints that use tokens

  #8035 opened Jun 3, 2025

• [Question]: why little json file parse slowly?

  #8034 opened Jun 3, 2025

• [Question]: Make ragflow as Dify's external Knowledge, the request is timeout

  #8030 opened Jun 3, 2025

• [Feature Request]: provide raw db migrate sqls for more restrict enterprise production environment

  #8029 opened Jun 3, 2025

• [Bug]: Can't upload more than 10 MB in batch, locally deployed docker

  #8027 opened Jun 3, 2025

• [Feature Request]: Knowledge base supporting multi-modal embedding models

  #8026 opened Jun 3, 2025

• [Bug]: 'Result window is too large, from + size must be less than or equal to: [10000] but was [12000]

  #8024 opened Jun 3, 2025

• [Question]: Chat doesn't retrieve answer from knowledge

  #8022 opened Jun 3, 2025

• [Question]: What is the situation when the user table is locked for a long time

  #8020 opened Jun 3, 2025

• [Question]:

  #8018 opened Jun 3, 2025

• [Feature Request]: local development doc is not good enough to install and startup the service on macOS

  #8014 opened Jun 3, 2025

• [Question]: Why is the document parsing progress not updated sometimes?

  #8011 opened Jun 3, 2025

• [Question]: When creating a new blank workflow, the "missing dsl" appears occasionally

  #8009 opened Jun 3, 2025

• [Question]: How to import and export the data for ragflow system

  #8008 opened Jun 3, 2025

• [Bug]: 22:19:42 [ERROR][Exception]: Exceptions from Trio nursery (1 sub-exception) -- **ERROR**: POST predict: Post "http://127.0.0.1:33395/completion": EOF

  #8007 opened Jun 2, 2025

• [Bug]: agent categorize node v0.18.0 is ok , but the v0.19.0 Can't inference the where the component input is. Please identify whose output is this component's input.

  #8006 opened Jun 2, 2025

• [Feature Request]: shall we update the Dockerfile and support to deploy web ui / server / task executor individually ?

  #8004 opened Jun 2, 2025

• [Question]: What context does the assistant (RagFlow chat) use when answering the question(s)

  #8003 opened Jun 2, 2025

• [Question]: Successful parsing and chunking but Retrieval Testing always has 0 Files

  #8001 opened Jun 2, 2025

• [Question]: no language set in 0.19 version?

  #7999 opened Jun 1, 2025

• [Feature Request]: Add support for parent-child chunk and chunk overlap functionality.

  #7996 opened May 31, 2025

• [Bug]: Memory leak ? after updating to v0.19.0

  #7995 opened May 31, 2025

• [Feature Request]:Optimize the current chunk split strategy based on separators and the max token setting.

  #7994 opened May 31, 2025

• [Bug]: WORD files could not be viewed in inference column.

  #7993 opened May 31, 2025

• [Bug]: Final answer repeats over and over again, without ends.

  #7992 opened May 31, 2025

• [Question]: I have created several assistants (chats) in the GUI. How can I get the "chat_it" list for all these chats?

  #7989 opened May 30, 2025

• [Bug]: upload file in Chat seems not working?

  #7988 opened May 30, 2025

• [Question]: Why can't I successfully invoke the rerank model deployed by vllm when using the API at /api/v1/retrieval?

  #7985 opened May 30, 2025

• [Question]: Is there a way to configure how particular extension should be parsed?

  #7984 opened May 30, 2025

• [Question]: v1/user/login

  #7982 opened May 30, 2025

• [Bug]: Use much more VRAM to chat with Ollama models than running in cmd

  #7981 opened May 30, 2025

• [Bug]: Unable to input Chinese in the chat box.

  #7976 opened May 30, 2025

• [Bug]: code compoent return error

  #7973 opened May 30, 2025

60 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Feat: Support production deployment mode by using gunicorn.

  #7845 commented on Jun 6, 2025 • 2 new comments

• [Question]: WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.

  #6071 commented on Jun 4, 2025 • 0 new comments

• ROADMAP 2025

  #4214 commented on Jun 4, 2025 • 0 new comments

• [Feature Request]: Increase small2big

  #7750 commented on Jun 4, 2025 • 0 new comments

• [Bug]: graphknowledge Error: Tree structure not found for treeKey: combo

  #7819 commented on Jun 4, 2025 • 0 new comments

• [Bug]: Adjusting the knowledge base of the file link will cause any file corresponding to the knowledge base that has already linked this file to need to be reparsed

  #7288 commented on Jun 4, 2025 • 0 new comments

• [Feature Request]: Add IMAP/POP3 retrieval

  #4014 commented on Jun 4, 2025 • 0 new comments

• [Question]: white screen for 3rd layer pages

  #7836 commented on Jun 5, 2025 • 0 new comments

• [Feature Request]: when will Langfuse support be added to the Agent? Is there a plan in the roadmap to implement a Q&A trace?

  #7901 commented on Jun 5, 2025 • 0 new comments

• [Feature Request]: Batch add chunks

  #4016 commented on Jun 5, 2025 • 0 new comments

• [Bug]: the knowledge graph is not displayed on the webpage

  #5099 commented on Jun 5, 2025 • 0 new comments

• [Question]: how can configure custom embedding models？

  #7863 commented on Jun 5, 2025 • 0 new comments

• [Feature Request]: deploy using WSGI.

  #1268 commented on Jun 5, 2025 • 0 new comments

• [Bug]: GraphRAG will not finish - multiple failures and parsing stuck after 36 hours and $200 on openai-mini models

  #7957 commented on Jun 5, 2025 • 0 new comments

• [Question]: Does ragflow support the self-built OceanBase database?

  #5350 commented on Jun 5, 2025 • 0 new comments

• [Question]: How to send requests to an MCP Server via HTTP？

  #7956 commented on Jun 5, 2025 • 0 new comments

• [Question]: Can document parsing support Mineru backend?

  #6791 commented on Jun 5, 2025 • 0 new comments

• [Question]: AUTH <password> called without any password configured for the default user. Are you sure your configuration is correct? Failed to launch ragflow service from source

  #5975 commented on Jun 5, 2025 • 0 new comments

• [Question]: http://es01:1200 is unhealthy for MacOS/M4

  #5038 commented on Jun 5, 2025 • 0 new comments

• [Feature Request]: Support Qdrant as vector DB in RagFlow

  #6546 commented on Jun 5, 2025 • 0 new comments

• [Bug]: ragflow-server can't connect to es01

  #6186 commented on Jun 5, 2025 • 0 new comments

• [Bug]: The “code” component is not available

  #7908 commented on Jun 5, 2025 • 0 new comments

• [Question]: How to login RAGFlow from another system.

  #7751 commented on Jun 6, 2025 • 0 new comments

• Is the current text-to-sql capable of outputting data analysis charts?

  #6732 commented on Jun 6, 2025 • 0 new comments

• [Feature Request]: The construction of the knowledge graph is extremely slow. How can I check its progress?

  #7959 commented on Jun 6, 2025 • 0 new comments

• [Question]: Slice symbol is not effective

  #7824 commented on Jun 6, 2025 • 0 new comments

• [Question]: Use the invoke component and prompt AssertionError('Please verify the input parameters first.')

  #7177 commented on Jun 6, 2025 • 0 new comments

• variables component agent

  #6333 commented on Jun 5, 2025 • 0 new comments

• Add a "total" field to All List APIs (HTTP & Python SDK)

  #7911 commented on May 30, 2025 • 0 new comments

• add schedule exec agent

  #7920 commented on Jun 5, 2025 • 0 new comments

• [Question]: How to create a flow that waits for multiple user messages before interacting?

  #7970 commented on May 30, 2025 • 0 new comments

• [Question]: why always `nothing found in knowledge base` ?

  #6742 commented on May 30, 2025 • 0 new comments

• [Feature Request]: Support dynamic update of Agent begin component parameters through interface

  #7953 commented on May 30, 2025 • 0 new comments

• [Question]: How to use MCP in Ragflow? Is there any documentation? If so, where can I find it?

  #7950 commented on May 30, 2025 • 0 new comments

• [Bug]: docker started mysql with an error

  #6794 commented on May 30, 2025 • 0 new comments

• [Question]:

  #7939 commented on May 30, 2025 • 0 new comments

• [Question]: The VRAM usage when running the model with ragflow is much higher than when running it with ollama run qwen3:14b.

  #7859 commented on May 30, 2025 • 0 new comments

• [Question]: Excuse me,is there an MCP service extracted from RAGFlow that I can use in Cherry Studio?

  #7913 commented on May 30, 2025 • 0 new comments

• [Question]: After I tagged the documents using the tag set, I found that the mixed similarity was greater than 100 during the retrieval test

  #7402 commented on May 30, 2025 • 0 new comments

• [Bug]: Error after chunking doucment: Inifinity is initing

  #7758 commented on Jun 3, 2025 • 0 new comments

• [Feature Request]: Prioritize multi language first.

  #3922 commented on Jun 3, 2025 • 0 new comments

• [Feature Request]: Function to implement own chunking / RAG methods

  #3937 commented on Jun 3, 2025 • 0 new comments

• [Feature Request]: Support external pdf parsers

  #3945 commented on Jun 3, 2025 • 0 new comments

• [Feature Request]: Can the ragflow project provide a rag technical document, or a paper?

  #3964 commented on Jun 3, 2025 • 0 new comments

• [Question]: How to add body info correctly in invoke component?

  #3990 commented on Jun 3, 2025 • 0 new comments

• [Feature Request]: Add GPUStack.ai as model provider

  #4064 commented on Jun 3, 2025 • 0 new comments

• [Feature Request]: Add (additional) Keywords and Questions to RAPTOR results

  #4104 commented on Jun 3, 2025 • 0 new comments

• [Bug]: Knowledge base deleted but bucket and documents in minio not deleted

  #4113 commented on Jun 3, 2025 • 0 new comments

• [Bug]: The knowledge is not used in my conversation.

  #7910 commented on Jun 3, 2025 • 0 new comments

• [Bug]: Embeddable Chat Iframe Causes Parent Page to Jump When New Messages Appear

  #7460 commented on Jun 3, 2025 • 0 new comments

• [Feature Request]: Outline KB integration (getoutline.com)

  #3665 commented on Jun 3, 2025 • 0 new comments

• [Feature Request]: When can the agent support the invocation of MCP services?

  #7936 commented on Jun 3, 2025 • 0 new comments

• [Feature Request]: It is suggested to parallelize the workflow nodes of the agent

  #7931 commented on Jun 3, 2025 • 0 new comments

• [Question]: got exception DoesNotExist What is the error?

  #6347 commented on Jun 3, 2025 • 0 new comments

• [Question]:when conducting a knowledge base search in the chat interface, it was found that ollama has a high cpu usage rate?

  #7881 commented on Jun 4, 2025 • 0 new comments

• [Question]: Why does Ragflow use Elasticsearch instead of Milvus?

  #2839 commented on Jun 4, 2025 • 0 new comments

• [Question]: Do you have plan to support other vector databases?

  #491 commented on Jun 4, 2025 • 0 new comments

• [Seeking help] Memory may be not released after PDF parsing

  #7299 commented on Jun 4, 2025 • 0 new comments

• [Bug]: Delimiter setting not working as intended

  #4704 commented on Jun 4, 2025 • 0 new comments

• [Bug]: 404 Not Found: The requested URL was not found on the server. If you entered the URL manually please check your spelling and try again.

  #5250 commented on Jun 4, 2025 • 0 new comments