1 Release published by 1 person

• 0.15.2 v0.15.2

  published Jan 22, 2025

72 Pull requests merged by 37 people

• fix qwen-vl blocking mode

  #13052 merged Jan 27, 2025

• fix: "Model schema not found" error only in agents (#12655)

  #12760 merged Jan 27, 2025

• fix:Improve MIME type detection for remote URL uploads using python-magic

  #12693 merged Jan 27, 2025

• fix: correct env vars for docker deployment

  #13055 merged Jan 27, 2025

• feat: add inner API to create workspace without requiring email

  #13021 merged Jan 26, 2025

• feat: docx image preview

  #13057 merged Jan 26, 2025

• fix: adjust iteration node dark style

  #13051 merged Jan 26, 2025

• add sign-content

  #13050 merged Jan 26, 2025

• Update Novita AI predefined models

  #13045 merged Jan 26, 2025

• fix: Unexpected tag creation when pressing enter during tag conversion

  #13041 merged Jan 25, 2025

• support bedrock kb: retrieve and generate

  #13027 merged Jan 25, 2025

• fix: can choose selected tools and show tool name instead of label

  #13025 merged Jan 24, 2025

• feat: add agent thinking content display of deepseek R1

  #12949 merged Jan 24, 2025

• fix(1.0): explore market page empty

  #13017 merged Jan 24, 2025

• chore: add Japanese translation: model_providers/bedrock

  #13016 merged Jan 24, 2025

• fix(1.0): update github plugin 404

  #13014 merged Jan 24, 2025

• Feat: dark mode of app configure

  #13010 merged Jan 24, 2025

• Change typo in translation

  #13004 merged Jan 24, 2025

• fix: wrong zh_Hans translation: Ohio

  #13006 merged Jan 24, 2025

• feat: add glm-4-air-0111

  #12997 merged Jan 24, 2025

• Feat/prompt editor dark theme

  #12976 merged Jan 23, 2025

• Plugin/merge main to plugin/beta 20250122

  #12962 merged Jan 23, 2025

• fix: plugin search api url

  #12977 merged Jan 23, 2025

• refactor: Update Firecrawl to use v1 API

  #12574 merged Jan 23, 2025

• fix: get property of string type variable caused page crash

  #12969 merged Jan 23, 2025

• fix: tools translate search

  #12950 merged Jan 22, 2025

• feat: add giteeAI risk control identification.

  #12946 merged Jan 22, 2025

• feat: add pg vector index

  #12338 merged Jan 22, 2025

• fix(1.0): invoke llm raise error

  #12753 merged Jan 22, 2025

• chore: update version to 0.15.2 in packaging and docker configurations

  #12940 merged Jan 22, 2025

• fix: add adapter for datasets update checking

  #12939 merged Jan 22, 2025

• feat(ark): support doubao 1.5 series of models

  #12935 merged Jan 22, 2025

• fix: resolve clipboard.writeText failure under HTTP protocol

  #12936 merged Jan 22, 2025

• fix: 500 error in Notion integration API

  #12934 merged Jan 22, 2025

• feat: workflow note dark theme

  #12932 merged Jan 22, 2025

• fix(batch_create_segment_to_index_task): count max_position in memory.

  #12929 merged Jan 22, 2025

• chore(build): bump poetry from 1.x to 2.x

  #12369 merged Jan 22, 2025

• add milvus full text search setting

  #12930 merged Jan 22, 2025

• fix: agent node output vars error

  #12931 merged Jan 22, 2025

• fix: Agent is configured for ReAct inference mode, an error is reported when viewing the agent log

  #12920 merged Jan 22, 2025

• fix: app token's last_used_at can't be updated when last_used_at is null

  #12770 merged Jan 22, 2025

• feat: support config chunk length by env

  #12925 merged Jan 22, 2025

• feat: add gemini-2.0-flash-thinking-exp-01-21

  #12924 merged Jan 22, 2025

• feat: Add ability to change profile avatar

  #12642 merged Jan 22, 2025

• fix:Fix a bug that returns null when the passed path is a file.

  #12775 merged Jan 22, 2025

• fix: change default PLUGIN_DAEMON_URL to http://localhost:5002

  #12915 merged Jan 21, 2025

• fix(1.0): add cross-env to pnpm run dev

  #12600 merged Jan 21, 2025

• fix update segment keyword with same content

  #12908 merged Jan 21, 2025

• bug: fixed bedrock rerank bug

  #12774 merged Jan 21, 2025

• chore: remove webapp ga

  #12909 merged Jan 21, 2025

• fix: variable panel scrollable

  #12769 merged Jan 21, 2025

• fix: OpenAI o1 Bad Request Error

  #12839 merged Jan 21, 2025

• Update deepseek model configuration

  #12899 merged Jan 21, 2025

• fix: external dataset hit test display issue(#12564)

  #12612 merged Jan 21, 2025

• add deepseek-reasoner

  #12898 merged Jan 21, 2025

• chore(fix): Invalid quotes for using Array[String] in HTTP request node as JSON body

  #12761 merged Jan 21, 2025

• fix: Issues related to the deletion of conversation_id (#12488)

  #12665 merged Jan 21, 2025

• chore(lint): fix quotes for f-string formatting by bumping ruff to 0.9.x

  #12702 merged Jan 21, 2025

• feat:Support Minimax-Text-01

  #12763 merged Jan 21, 2025

• fix: correct validation for agent node which is invoked before publishing the app

  #12805 merged Jan 21, 2025

• fix(1.0): unexpected error raise

  #12812 merged Jan 21, 2025

• chore(1.0): improve some environment variables

  #12814 merged Jan 21, 2025

• fix(1.0): sometimes add tool raise error

  #12821 merged Jan 21, 2025

• fix: serply credential check query might return empty records

  #12784 merged Jan 21, 2025

• feat: allow updating chunk settings for the existing documents

  #12833 merged Jan 21, 2025

• fix: SparkLite API Auth error (#12781)

  #12790 merged Jan 20, 2025

• fix: "parmas" spelling mistake.

  #12875 merged Jan 20, 2025

• Fix suggested_question_prompt

  #12738 merged Jan 20, 2025

• fix(i18n): correct typo in zh-Hant translation

  #12852 merged Jan 20, 2025

• chore: fix chinese translation for 'recall'

  #12772 merged Jan 20, 2025

• fix: DeepSeek API Error with response format active (text and json_object)

  #12747 merged Jan 20, 2025

• feat: enhance credential extraction logic based on configurate method

  #12853 merged Jan 20, 2025

9 Pull requests opened by 8 people

• fix: run out of memory due to indexing/embedding documents

  #12882 opened Jan 20, 2025

• fix document extractor node incorrectly processing doc and ppt files

  #12902 opened Jan 21, 2025

• fix: embedding demension change bug

  #12914 opened Jan 21, 2025

• [FIX]Ruff: lint errors for E731

  #13018 opened Jan 24, 2025

• ComfyUI Tool realtive WS paths

  #13029 opened Jan 24, 2025

• Feat: add gpustack image tools

  #13042 opened Jan 25, 2025

• refactor: avoid to use extra space when finding model by name

  #13043 opened Jan 25, 2025

• fix: fix a typo of get_customizable_model_schema method name

  #13044 opened Jan 25, 2025

• Feature/new workflow loop node

  #13058 opened Jan 26, 2025

156 Issues closed by 23 people

• There is a problem with clicking on the answer link in markdown format

  #12954 closed Jan 27, 2025

• /chat-messages API handle VL model input error

  #12975 closed Jan 27, 2025

• "Model schema not found" error only in agents

  #12655 closed Jan 27, 2025

• Incorrect File Format Returned for .docx Upload via Remote URL

  #12551 closed Jan 27, 2025

• Error occurred when invoking the deepseek series of models.

  #13067 closed Jan 27, 2025

• need conversation_id to the parameters returned by the extension tool API

  #11556 closed Jan 26, 2025

• Automatically load documents from one directory.

  #11503 closed Jan 26, 2025

• Update Novita AI predefined models

  #13049 closed Jan 26, 2025

• Deepseek integration fail.

  #13046 closed Jan 26, 2025

• Request to Add API Base Configuration for Google Gemini Model Provider

  #13048 closed Jan 25, 2025

• When two or more nodes are connected to multiple classes of a question classifier, only the last one configured will run in parallel.

  #11528 closed Jan 25, 2025

• Optimize ReAct Generation: Remove Implicit "continue" and Directly Append "Thought:"

  #11525 closed Jan 25, 2025

• When testing the custom tool, the historical url was used

  #11526 closed Jan 25, 2025

• Content not equal within a shot time gap while using subgraph (sub workflow) in workflow

  #11517 closed Jan 25, 2025

• feat: add VTT data transform to Document extractor

  #11515 closed Jan 25, 2025

• [xinference] Error: An error occurred during streaming

  #8253 closed Jan 25, 2025

• Unexpected tag creation when pressing enter during tag conversion

  #13039 closed Jan 25, 2025

• Bedrock introduced a new feature: retrieve and generate

  #13026 closed Jan 25, 2025

• Click Setting up an admin account but back to login page immediately

  #13024 closed Jan 25, 2025

• when I view a document on page 3, it always jumps to the first page after returning, which is supposed to stay on page 3

  #11488 closed Jan 24, 2025

• Add an ASR Model Provider Interface for Alibaba Cloud

  #11462 closed Jan 24, 2025

• same text chunk with different embedding model get exactly same vector hash value

  #11464 closed Jan 24, 2025

• Agent side of conversation vanishing

  #11408 closed Jan 24, 2025

• Can topK > 10 in knowledge retrieval node?

  #11469 closed Jan 24, 2025

• Workflow can not be published if it contains Agent node

  #12804 closed Jan 24, 2025

• Added support for including the reasoning process in the output of Reasoning Model.

  #12891 closed Jan 24, 2025

• will dify optimise docker compose to local development

  #13011 closed Jan 24, 2025

• The API version for the o1 series models of Azure OpenAI is unavailable.

  #13012 closed Jan 24, 2025

• Reasoning models

  #12917 closed Jan 24, 2025

• [ReOpen] Segmentation Recognition Fails When Using create_by_text in Parent-Child Mode in the Knowledge Base API

  #13013 closed Jan 24, 2025

• latest docker version bug?

  #13001 closed Jan 24, 2025

• Segmentation Recognition Fails When Using create_by_text in Parent-Child Mode in the Knowledge Base API

  #13007 closed Jan 24, 2025

• Installation tutorial without Docker?

  #13003 closed Jan 24, 2025

• When I set up an administrator account, an Internal Server Error occurred.

  #12985 closed Jan 24, 2025

• add glm4 new models：glm-4-air-0111

  #12864 closed Jan 24, 2025

• what version of torch，tensorflow and flax should be used? Dify version is 0.15.1

  #12996 closed Jan 24, 2025

• Segment Maximum Length is sometimes limited to 1000 and at other times can be set to 4000

  #12995 closed Jan 24, 2025

• A large amount of data in the knowledge base is in indexed and queued states

  #12986 closed Jan 23, 2025

• Question: After building a Dify workflow, how can I ensure that the next query is not influenced by the previous query history? How should I modify this in the interface? Thanks!

  #12983 closed Jan 23, 2025

• Free YouTube Transcript API failing to fetch transcript in Dify Cloud

  #12978 closed Jan 23, 2025

• Output of the Prompt Generator is Truncated, Incomplete. How to Set the Maximum Output Token for the Prompt Generator?

  #12980 closed Jan 23, 2025

• docker-compose up -d failed

  #12972 closed Jan 23, 2025

• The Chatflow "Template" node implements the interactive button functionality

  #12973 closed Jan 23, 2025

• 'App not found.' Message Displayed Despite Successful App Creation"

  #12078 closed Jan 23, 2025

• The LLM Node cannot read file with a custom field type of "File" when using a multimodal model.

  #11655 closed Jan 23, 2025

• Ability to launch Chatflow via Webhook

  #12830 closed Jan 23, 2025

• [Bug] alter model from gpt3 to wenxin api, the flow fails at first node requires model

  #10281 closed Jan 23, 2025

• In the workflow’s code executor, an exception occurs when the value of the input parameter is set to the string ‘daily’, however others strings are acceptable.

  #12968 closed Jan 23, 2025

• Move Firecrawl from v0 to v1 API

  #12571 closed Jan 23, 2025

• workflow task support multi ouput

  #12970 closed Jan 23, 2025

• The HTTP request cannot access HTTPS and prompts a 403 error.

  #12965 closed Jan 23, 2025

• Application error: a client-side exception has occurred (see the bprowser console for more information).

  #12966 closed Jan 23, 2025

• Firecrawl Local Hosted Issue: Line 1 column 1 (char 0)

  #12961 closed Jan 22, 2025

• Dify document stuck at Queue

  #12959 closed Jan 22, 2025

• `poetry.lock` requires regeneration when install locally from source code

  #12629 closed Jan 22, 2025

• The search of tools does not support Chinese.

  #12762 closed Jan 22, 2025

• GiteeAI tool adds risk control identification

  #12945 closed Jan 22, 2025

• I can't set the OpenAI config in model marktet

  #12958 closed Jan 22, 2025

• I deployed the MiniCPM-o-2.6:7.6b model using Ollama, but the output interpretation of the image content was off-topic.

  #12955 closed Jan 22, 2025

• Docker version model list is missing/incorrect

  #12942 closed Jan 22, 2025

• Deploy Sandbox From Source

  #12947 closed Jan 22, 2025

• feat(vdb): add HNSW vector index for PG vector store

  #12341 closed Jan 22, 2025

• 适合CPU（arm）架构的dify镜像

  #12943 closed Jan 22, 2025

• 0.15.2 Issue Tracker

  #12927 closed Jan 22, 2025

• 'Nonetype' object has no attribute 'get'

  #12679 closed Jan 22, 2025

• The agent reports an error using the gpt-4o-mini model

  #12881 closed Jan 22, 2025

• When installing the Fish Audio plugin, an "invalid plugin id" message will appear at the top.

  #12923 closed Jan 22, 2025

• support doubao 1.5 series of models

  #12938 closed Jan 22, 2025

• Copy API curl request error

  #12818 closed Jan 22, 2025

• Bug: 500 Error in Notion Integration API

  #12933 closed Jan 22, 2025

• import segments from cvs:can't adapt type 'dict'

  #12717 closed Jan 22, 2025

• bump poetry to 2.x

  #12368 closed Jan 22, 2025

• Retrieval Setting： Search method configuration problem

  #12878 closed Jan 22, 2025

• Agent is configured for ReAct inference mode, an error is reported when viewing the agent log

  #12919 closed Jan 22, 2025

• API Secret key `LAST USED` can not be update

  #12911 closed Jan 22, 2025

• The chunk size of the knowledge base cannot be modified through environment variables.

  #12500 closed Jan 22, 2025

• [Feature request]New Gemini Model

  #12918 closed Jan 22, 2025

• Add ability to customize user avatars

  #12641 closed Jan 22, 2025

• The "Create from Blank" page cannot load static files

  #12912 closed Jan 22, 2025

• Add QwQ-32B-Preview support

  #11260 closed Jan 22, 2025

• Refactor Python SDK and Construct dify_oapi to Improve Development Efficiency

  #10238 closed Jan 21, 2025

• Sagemaker Can't load model name

  #10289 closed Jan 21, 2025

• Knowledge Retrieval with external knowledge api response metadata

  #11422 closed Jan 21, 2025

• The custom segment length cannot exceed 1000 anymore?

  #12916 closed Jan 21, 2025

• Manually Added Keyword Ineffective in RAG Keyword Retrieval

  #12884 closed Jan 21, 2025

• After using the dify tutorial docker deployed on the server, the page cannot be displayed normally and an api error is displayed

  #12907 closed Jan 21, 2025

• Add AWS Bedrock rerank model invalid param

  #12640 closed Jan 21, 2025

• Setup AWS Bedrock with blank AKSK returns invalid_param error after upgrading to 0.14.1

  #11819 closed Jan 21, 2025

• Bug Report: Issue with Dify AI Version 0.14.1 - OpenAI Bad Request Error (400)

  #11981 closed Jan 21, 2025

• Why does my agent robot reply not in typewriter-like flow mode but in blocks in blocking mode

  #12904 closed Jan 21, 2025

• There is a problem with clicking on the answer link in markdown format

  #12903 closed Jan 21, 2025

• Test external knowledge base retrieval, the console reports an error in v0.15.0

  #12564 closed Jan 21, 2025

• When entering an excel document and extracting the document, if there are multiple sheets in excel, other sheets cannot be read

  #12895 closed Jan 21, 2025

• Add DeepSeek R1 model

  #12890 closed Jan 21, 2025

• Workflow 403 Forbidden

  #12896 closed Jan 21, 2025

• Retrieval Test of external knowledge error

  #12741 closed Jan 21, 2025

• Invalid JSON string when using Array[String] in HTTP request node

  #12759 closed Jan 21, 2025

• Issues related to the deletion of conversation_id

  #12488 closed Jan 21, 2025

• "parent-child" mod don't works on txt documents

  #12819 closed Jan 21, 2025

• open 127.0.0.1/install Internal Server Error

  #12892 closed Jan 21, 2025

• update ruff to 0.9.0 to conform 2025 style guides

  #12703 closed Jan 21, 2025

• The retry mechanism in the workflow is not working.

  #12746 closed Jan 21, 2025

• web page lack full text search and hybrid search

  #12749 closed Jan 21, 2025

• Fail to add multi-model Qwen2-vl

  #12750 closed Jan 21, 2025

• Issue: Delayed Access and Login Problems After Deployment on Oracle Cloud

  #11833 closed Jan 21, 2025

• serply credential check query might return empty records

  #12783 closed Jan 21, 2025

• HTTP Request with sys.file in Form-Data Results in Empty Files

  #12752 closed Jan 21, 2025

• knowledgebase retrieval jina timeout error

  #12754 closed Jan 21, 2025

• variables `Iteration item` and `Iteration index` cannot be displayed correctly which used HTTP node in Iteration node.

  #12765 closed Jan 21, 2025

• When the Agent's model is set to DeepSeek V3 Chat, the tool invocation count becomes abnormal.

  #12778 closed Jan 21, 2025

• [PANIC]failed to install python dependencies: exit status 1

  #12786 closed Jan 21, 2025

• How to print infomation in code execution nodes of the workflow?

  #12789 closed Jan 21, 2025

• change and save Chunk Settings

  #12794 closed Jan 21, 2025

• chat assistant repeat answer same contents

  #12823 closed Jan 21, 2025

• Note: NumExpr detected 30 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.

  #12857 closed Jan 21, 2025

• Add an option for the user side to automatically speak out.

  #11409 closed Jan 20, 2025

• Running the preset project Long Story Generator (Iteration) reports an error

  #11361 closed Jan 20, 2025

• I think the model configuration should support changing the alias of the model instead of the real model name

  #11279 closed Jan 20, 2025

• The output text will be prefixed with escape characters.

  #11259 closed Jan 20, 2025

• add github model as model provider

  #11329 closed Jan 20, 2025

• Question about chroma vector database implementation.

  #11391 closed Jan 20, 2025

• A tool used to retrieval some info from LangFuse(self-hosted)

  #11040 closed Jan 20, 2025

• BUG when importing DSL project file with TOOLs [CHATFLOW]

  #10911 closed Jan 20, 2025

• can't connect an audio file in speech to text tool

  #11046 closed Jan 20, 2025

• Large input content will lead to an stuck error on user conversation page and backend conversation management page.

  #11250 closed Jan 20, 2025

• Return Full Article with Highlighted Matches for Local Knowledge Base Queries

  #11185 closed Jan 20, 2025

• The session style has issues after Dify is embedded in the website.

  #11205 closed Jan 20, 2025

• Is it possible to add the function of left folding in the exploration column?

  #10850 closed Jan 20, 2025

• enhancement for chatbot dialog width

  #10843 closed Jan 20, 2025

• Add Baichuan4-Turbo/Baichuan4-Air model in Baichuan AI Provider

  #10773 closed Jan 20, 2025

• Allow Modifying sys.user_id Variables Within Workflow Conversation

  #10721 closed Jan 20, 2025

• ‌‌‌‌‌OpenRouter, o1-mini and o1-preview error

  #10279 closed Jan 20, 2025

• In Webapp conversation, do not scroll down the window while a long response is being generated

  #9884 closed Jan 20, 2025

• SparkLite API Auth error

  #12781 closed Jan 20, 2025

• Common i18n issue

  #12883 closed Jan 20, 2025

• How can I change it so that only the content returned by LLM is returned in the form of streaming?

  #12826 closed Jan 20, 2025

• An error is reported when the Array[String] parameter of dify is used as http json

  #12809 closed Jan 20, 2025

• deepseek response format error

  #12868 closed Jan 20, 2025

• DeepSeek API Error with response format active (text and json_object)

  #12713 closed Jan 20, 2025

• Full-text search is not supported in current Milvus version (requires >= 2.5.0)

  #12837 closed Jan 20, 2025

• API key error occurs when setting model load balancing with Google model

  #12850 closed Jan 20, 2025

• Can the input length limit in the recall test of the knowledge base be modified?

  #12855 closed Jan 20, 2025

• Use /datasets/{dataset_id}/document/create-by-text API error

  #12863 closed Jan 20, 2025

• How to get request url parameters in the Published apps？

  #12811 closed Jan 20, 2025

• upload file bug

  #12748 closed Jan 20, 2025

• Knowledge more than 4 in app,then the answer can not be complete

  #12755 closed Jan 20, 2025

• Internal Server Error

  #12756 closed Jan 20, 2025

• When retrieving from the Ragflow external Q/A knowledge base, the returned content only includes the questions, without the answers.

  #12684 closed Jan 20, 2025

• Add "Go to Page N" Functionality in Log Pagination

  #12865 closed Jan 20, 2025

• dify worker stuck

  #12876 closed Jan 20, 2025

• hit testing giving time out error

  #12880 closed Jan 20, 2025

• Dify采用docker-compose方式部署（遇到weaviate组件缺失问题）

  #12872 closed Jan 20, 2025

• 手动添加的标签召回问题

  #12845 closed Jan 20, 2025

• The return result of the Feishu document creation tool includes a field named json, which seems like a strange name. 飞书文档创建工具的返回结果中有 json这个字段，这样的名字很奇怪

  #12873 closed Jan 20, 2025

• The return result of the Feishu document creation tool includes a field named json, which seems like a strange name. 飞书文档创建工具的返回结果中有 json这个字段，这样的名字很奇怪

  #12874 closed Jan 20, 2025

• cors bug!!!

  #12866 closed Jan 20, 2025

50 Issues opened by 44 people

• Celery worker and beat fail when LOG_LEVEL is unset

  #13071 opened Jan 27, 2025

• Incorrect protocol and port detection when APP_WEB_URL is unset behind Nginx

  #13070 opened Jan 27, 2025

• API DeepSeek json_object empty response 0.15.2

  #13068 opened Jan 27, 2025

• Document Embedding issues while using sagemaker model

  #13064 opened Jan 26, 2025

• Playwright in Code Block

  #13063 opened Jan 26, 2025

• Relative WS path for comfyui

  #13061 opened Jan 26, 2025

• Enhance Plugin Marketplace Usability: Add 'Installed' Tag for Better Visibility

  #13054 opened Jan 26, 2025

• File Upload Failures Persist Until NGINX_CLIENT_MAX_BODY_SIZE Is Adjusted Alongside .env Configuration

  #13053 opened Jan 26, 2025

• Request to Add API Base Configuration for Google Gemini Model Provider

  #13047 opened Jan 25, 2025

• Knowledge -> Sync from Notion ：DOM display error when custom icons are present

  #13040 opened Jan 25, 2025

• feat: image tools for gpustack

  #13038 opened Jan 25, 2025

• Enhance TLS Security: Enforce Minimum TLS Version for starttls()

  #13037 opened Jan 25, 2025

• [openai] Bad Request Error, Error code: 404 - {'error': {'message': 'The model o1-2024-12-17 does not exist or you do not have access to it.', 'type': 'invalid_request_error', 'param': None, 'code': 'model_not_found'}}

  #13036 opened Jan 25, 2025

• The stop response api is ineffective for "chat assistant" applications but works for "ChatFlow-type" applications.

  #13035 opened Jan 25, 2025

• Add tool support for liblib to generate image

  #13033 opened Jan 25, 2025

• Integrating <WorkflowContainer /> into my own webpage

  #13031 opened Jan 25, 2025

• Agent Node Displays Default Maximum Iterations as 3 but Triggers 'Cannot Be Empty'

  #13030 opened Jan 25, 2025

• Reset the embedded chat

  #13023 opened Jan 24, 2025

• Linter Rule Tracker for Jan, 2025

  #13022 opened Jan 24, 2025

• Firecrawl extraction doesn't apply "only_main_content" parameter when set to true in v0.15.2

  #13019 opened Jan 24, 2025

• v1.0.0-beta.1 agent doesn't work with OpenAI-API-compatible plugin in workflow

  #13008 opened Jan 24, 2025

• Using Template Node create a login form, can be configured to not display in the conversation when the Button is pressed.

  #12998 opened Jan 23, 2025

• Database Size Growth Issue: Workflow Execution Logs and Retention Policy

  #12991 opened Jan 23, 2025

• Support grounding for Vertex Gemini

  #12990 opened Jan 23, 2025

• Regarding the functional expansion to enable the use of the o1 model in Azure Open AI.

  #12988 opened Jan 23, 2025

• Parameters are required to control the timeout of ollama requests.

  #12987 opened Jan 23, 2025

• sandbox container access https request prompt 403 error, but api container access https request is OK

  #12974 opened Jan 23, 2025

• Upload knowledge base error, error information:Meta endpoint! Unexpected status code: 502, with response body: None.

  #12971 opened Jan 23, 2025

• dify cannot create langsmith project

  #12967 opened Jan 23, 2025

• Custom values via embedded chat

  #12963 opened Jan 22, 2025

• Vector database connection error

  #12960 opened Jan 22, 2025

• Please assist in integrating the voyage embedding and rerank model into the plugin.

  #12956 opened Jan 22, 2025

• Chatflow LLM node unwantedly include sys.files in LLM requests

  #12952 opened Jan 22, 2025

• LLM Node in a Chatflow Didn't Work Normally

  #12944 opened Jan 22, 2025

• add default value for selected variable in the input field

  #12937 opened Jan 22, 2025

• support customize AWS Bedrock endpoint URL

  #12926 opened Jan 22, 2025

• Could the RAG strategy also be presented in the form of a plugin?

  #12921 opened Jan 22, 2025

• embedding demension change bug

  #12913 opened Jan 21, 2025

• About the Development Plan of Dify Desktop Application

  #12910 opened Jan 21, 2025

• Feature Request: Configurable Concurrency for Workflow Batch Runs

  #12906 opened Jan 21, 2025

• Support `NGINX_ENABLED` Environment Variable

  #12905 opened Jan 21, 2025

• Lark suite aka Feishu - access more variables ouput

  #12900 opened Jan 21, 2025

• uploaded source files can be reviewed and downloaded directly on the webpage.

  #12897 opened Jan 21, 2025

• Potential Risk of Hitting Milvus Collection Limit in Dify Design

  #12894 opened Jan 21, 2025

• huge csv file occupies huge memory while indexing document

  #12893 opened Jan 21, 2025

• Document extractor node incorrectly handles doc and ppt files

  #12889 opened Jan 21, 2025

• Define or import, publish and use Workflows at runtime without the frontend

  #12888 opened Jan 20, 2025

• Podcast Generator extend TTS Compability options.

  #12887 opened Jan 20, 2025

• add variable with type equals to 'array[object]' in the input field

  #12877 opened Jan 20, 2025

• Clarification of software architecture, namely pluginDaemon

  #12869 opened Jan 20, 2025

141 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/add linkup

  #11854 commented on Jan 22, 2025 • 10 new comments

• Introduce Configurable Upper-Limits/default values for max retries and retry Interval in Error Handling(#12787)

  #12788 commented on Jan 22, 2025 • 3 new comments

• resolved: add basic auth for chatbot , implemented basic authentication after deploying with Docker Compose

  #10215 commented on Jan 22, 2025 • 2 new comments

• The JS code doesn't work, but Python does

  #12287 commented on Jan 23, 2025 • 0 new comments

• Internal Server Error

  #12664 commented on Jan 23, 2025 • 0 new comments

• Implement OpenAI's Realtime API

  #8986 commented on Jan 23, 2025 • 0 new comments

• Can't locate tools when importing workflows from dev environment to production environment.

  #11964 commented on Jan 23, 2025 • 0 new comments

• "Knowledge" block settings: The implementation and the documentation are different.

  #11976 commented on Jan 23, 2025 • 0 new comments

• Support for multiple servers in custom tools OpenAPI definition

  #10999 commented on Jan 23, 2025 • 0 new comments

• add Ollama rerank model

  #10917 commented on Jan 23, 2025 • 0 new comments

• Agent response parse error

  #11978 commented on Jan 23, 2025 • 0 new comments

• Auth token never refreshes - keep getting logged out in the middle of a session

  #11973 commented on Jan 23, 2025 • 0 new comments

• Dify Cloud quota for OpenAI can't be used

  #11959 commented on Jan 23, 2025 • 0 new comments

• Support for Retrieving Streamed Output from Any Model Node in Dify's Workflow API

  #12009 commented on Jan 23, 2025 • 0 new comments

• Environment specific tool specs when using Dify via its API SDK

  #11816 commented on Jan 23, 2025 • 0 new comments

• Creating Knowledge Base Error

  #12532 commented on Jan 24, 2025 • 0 new comments

• DeepSeek Load Balancing Bug

  #11160 commented on Jan 24, 2025 • 0 new comments

• Parallel group execution sequence error

  #12848 commented on Jan 24, 2025 • 0 new comments

• Slow Embedding Performance with External Model API Calls in Dify - Need for Concurrency Adjustment

  #12032 commented on Jan 24, 2025 • 0 new comments

• feat: replace default json to orjson for better performance

  #12028 commented on Jan 24, 2025 • 0 new comments

• Integration of Quantum Computation Node with AWS Braket

  #11663 commented on Jan 20, 2025 • 0 new comments

• Feature Request: Add Support for Arabic Language in "Segmenting in Question & Answer Format"

  #10811 commented on Jan 22, 2025 • 0 new comments

• Error uploading files while building the knowledge base

  #11657 commented on Jan 22, 2025 • 0 new comments

• Permanently corrupted knowledge base where subsequent file uploads fail to index

  #11958 commented on Jan 22, 2025 • 0 new comments

• Display conversation message feedbacks on the UI

  #11969 commented on Jan 22, 2025 • 0 new comments

• The chat api supports base64 image files

  #9430 commented on Jan 22, 2025 • 0 new comments

• [Embedded chatbot] GET input params passed in iframe src are lost after clicking "Reset Conversation"

  #11025 commented on Jan 22, 2025 • 0 new comments

• Image url will not completed if a document start with an image

  #11791 commented on Jan 22, 2025 • 0 new comments

• Does Dify have plans to support rerank and embedding models launched by vLLM?

  #11857 commented on Jan 22, 2025 • 0 new comments

• Workflow Schedule Feature

  #9168 commented on Jan 22, 2025 • 0 new comments

• Invalid text content type: <class 'NoneType'>. Expected str.

  #10724 commented on Jan 22, 2025 • 0 new comments

• Occasional error "operation not permitted" for Code Blocks

  #11227 commented on Jan 22, 2025 • 0 new comments

• JS Code Node: "operation not permitted"

  #11674 commented on Jan 22, 2025 • 0 new comments

• Add New Embedding Model to Voyage

  #11744 commented on Jan 23, 2025 • 0 new comments

• Knowledge base search supports Filtered Search

  #12796 commented on Jan 23, 2025 • 0 new comments

• Cannot open new or existing application with old browser

  #12030 commented on Jan 23, 2025 • 0 new comments

• missing argument retriever_resources in api response

  #12479 commented on Jan 23, 2025 • 0 new comments

• environment variable is null when execute node

  #10065 commented on Jan 27, 2025 • 0 new comments

• Ensure the 'inputs' field in /chat-messages takes effect every time

  #7955 commented on Jan 21, 2025 • 0 new comments

• Make Dify compatible with MySQL database

  #8364 commented on Jan 22, 2025 • 0 new comments

• feat: add support for Azure speech

  #11375 commented on Jan 22, 2025 • 0 new comments

• Add support for ModelScope community

  #11397 commented on Jan 24, 2025 • 0 new comments

• feat: parallel delete vector

  #11720 commented on Jan 22, 2025 • 0 new comments

• Feature/sambanovacloud integration dify

  #11918 commented on Jan 23, 2025 • 0 new comments

• Fixed Several Exceptions in Workflow Usage

  #12075 commented on Jan 26, 2025 • 0 new comments

• Add "Scroll Up to Load More" and Fix Old Replies Issue in Chat Component.

  #12348 commented on Jan 26, 2025 • 0 new comments

• Fix: new upgrade page

  #12417 commented on Jan 23, 2025 • 0 new comments

• Fix the problem of Workflow terminates after parallel tasks execution, merge node not triggered

  #12498 commented on Jan 21, 2025 • 0 new comments

• Fix bug large data no render

  #12683 commented on Jan 20, 2025 • 0 new comments

• fix: fix the formatter is not applied on log file

  #12704 commented on Jan 21, 2025 • 0 new comments

• fix: ignore tables like embedding_vector_index_***_nod when migrate

  #12776 commented on Jan 21, 2025 • 0 new comments

• feat: bedrock_endpoint_url

  #12838 commented on Jan 27, 2025 • 0 new comments

• Add support for import custom Q&A file

  #12844 commented on Jan 21, 2025 • 0 new comments

• Enhancement Podcast Generator tool to support all TTS models

  #12858 commented on Jan 25, 2025 • 0 new comments

• i want the http request node support sse

  #12004 commented on Jan 24, 2025 • 0 new comments

• Likes and dislikes are not shown in chat logs, but present in db

  #11793 commented on Jan 24, 2025 • 0 new comments

• If a parallel node in a workflow contains if/else nodes, if the node after the parallel node does not wait for all parallel nodes to complete

  #11756 commented on Jan 24, 2025 • 0 new comments

• Some Exceptions Occurring with Workflow Usage

  #12074 commented on Jan 25, 2025 • 0 new comments

• Using MAKE.com workflow automation: bad_request: The browser (or proxy) sent a request that this server could not understand

  #12103 commented on Jan 25, 2025 • 0 new comments

• When running Alibaba's OCR sdk in the workflow, an error message appears: "error: operation not permitted"

  #11873 commented on Jan 25, 2025 • 0 new comments

• When updating in ChatFlow, it will automatically restore, and sometimes the page will flash

  #11874 commented on Jan 25, 2025 • 0 new comments

• Asynchronous Processing in Workflows?

  #11805 commented on Jan 25, 2025 • 0 new comments

• Request for Support of KAG (Knowledge-Aware Graph)

  #12085 commented on Jan 25, 2025 • 0 new comments

• Add PDF Processing Support for Vertex AI Gemini Provider with Graph Recognition

  #12069 commented on Jan 25, 2025 • 0 new comments

• Need a new provider for Tongyi international users

  #11970 commented on Jan 26, 2025 • 0 new comments

• Agent NEVER get any result from workflow tool

  #12108 commented on Jan 26, 2025 • 0 new comments

• When receiving binary files response in HTTP Request Node, it wouldn't decode Base64Encode. I need it.

  #12136 commented on Jan 26, 2025 • 0 new comments

• The completion-messages interface returns an exception

  #12111 commented on Jan 26, 2025 • 0 new comments

• Regarding the issue of Firecrawl unrestricted crawling when creating a knowledge base.

  #11029 commented on Jan 26, 2025 • 0 new comments

• Can the minimum threshold for knowledge base annotation be lowered?

  #11748 commented on Jan 26, 2025 • 0 new comments

• Pending tools and model runtimes

  #11588 commented on Jan 27, 2025 • 0 new comments

• support qwen long input format：system role input twice

  #11820 commented on Jan 20, 2025 • 0 new comments

• preview with data from logs

  #11618 commented on Jan 20, 2025 • 0 new comments

• return page number of word docx documents upon retrieval

  #11891 commented on Jan 20, 2025 • 0 new comments

• Enhance debugging target node in a workflow with a shortcut to use inputs from last test log

  #11826 commented on Jan 20, 2025 • 0 new comments

• Improve Font Color Adaptability in Chatbot Theme Settings

  #11632 commented on Jan 20, 2025 • 0 new comments

• Add Upward Scrolling to Load More Messages in Chat Component

  #11848 commented on Jan 20, 2025 • 0 new comments

• feat: Remove 'url' from File structure in the logs

  #11895 commented on Jan 21, 2025 • 0 new comments

• Can we add batch delete annotations function?

  #12836 commented on Jan 21, 2025 • 0 new comments

• The query page of the knowledge base, will automatically slide up and down, see the picture for details

  #12758 commented on Jan 21, 2025 • 0 new comments

• I found an abnormal bug while using HTTP to request nodes

  #12792 commented on Jan 21, 2025 • 0 new comments

• Support Object type in the custom tool via OpenAPI.

  #9025 commented on Jan 21, 2025 • 0 new comments

• error when using the "http request" node, with the uploaded file larger than 900KB

  #12216 commented on Jan 21, 2025 • 0 new comments

• Workflow terminates after parallel tasks execution, merge node not triggered

  #12492 commented on Jan 21, 2025 • 0 new comments

• Add support for import some Q&A text not need LLM to generate results

  #12842 commented on Jan 21, 2025 • 0 new comments

• Introduce Configurable Upper-Limits/default values for max retries and retry Interval in Error Handling

  #12787 commented on Jan 21, 2025 • 0 new comments

• Add Export Functionality to the Knowledge Base Module

  #12862 commented on Jan 21, 2025 • 0 new comments

• Connection and variables are lost when block changes are made

  #11912 commented on Jan 21, 2025 • 0 new comments

• [Feature] Allows the workflow to handle based on user's feedback

  #12859 commented on Jan 20, 2025 • 0 new comments

• conversation detail api is not allowed in cloud version

  #12663 commented on Jan 20, 2025 • 0 new comments

• workflow node http-request header cover

  #12841 commented on Jan 20, 2025 • 0 new comments

• The SDK cannot access locally deployed services and cannot specify a base_url to localhost.

  #11664 commented on Jan 20, 2025 • 0 new comments

• Add multiple selection function to the segmented list of documents in the knowledge base

  #11730 commented on Jan 20, 2025 • 0 new comments

• using a function call to invoke the model reported 'Extra inputs are not permitted'.

  #11795 commented on Jan 20, 2025 • 0 new comments

• [Chatflow] sys.files does not contain a property to use in a list-operator block to differenciate if it was a local upload or upload via URL

  #11817 commented on Jan 20, 2025 • 0 new comments

• add error handler for parameter extractor

  #11815 commented on Jan 20, 2025 • 0 new comments

• Add Method to Retrieve Conversation by ID in Dify Python Client

  #11532 commented on Jan 20, 2025 • 0 new comments

• Next step: do not add connections

  #11824 commented on Jan 20, 2025 • 0 new comments

• Support code knowledge base

  #11890 commented on Jan 20, 2025 • 0 new comments

• I use a knowledge base to parse documents into QA data. Why can't I use it

  #11802 commented on Jan 20, 2025 • 0 new comments

• Addition of "Regex Splitter" for Text Chunking in Dify

  #11661 commented on Jan 20, 2025 • 0 new comments

• Tools to Enhance GitHub Integration: Add PR Creation, Code Commit, Merge, and Extraction Tools

  #11757 commented on Jan 20, 2025 • 0 new comments

• The problem classifier integrates Claude's arbitrary model, which generates errors during execution and is incompatible;

  #11642 commented on Jan 20, 2025 • 0 new comments

• Customizable Tag-Based Filtering for Applications Display in Discovery Sidebar

  #11904 commented on Jan 20, 2025 • 0 new comments

• I have a requirement to add several models that have different API addresses but the same name. However, I cannot add them because there is no option for aliases/alternate names.

  #11634 commented on Jan 20, 2025 • 0 new comments

• Add data import and export functions

  #11870 commented on Jan 20, 2025 • 0 new comments

• I want to control the Knowledge items in the Knowledge Retrieval Base of the Workflow via an API.

  #11900 commented on Jan 21, 2025 • 0 new comments

• Add SambaNova support

  #11615 commented on Jan 21, 2025 • 0 new comments

• Integrate Prompt Versioning in Dify's LLM Block

  #10960 commented on Jan 21, 2025 • 0 new comments

• Add default values to user-defined input field

  #11290 commented on Jan 21, 2025 • 0 new comments

• Failed to format qa document

  #12582 commented on Jan 22, 2025 • 0 new comments

• add basic auth for chatbot , implemented basic authentication after deploying with Docker Compose

  #10475 commented on Jan 22, 2025 • 0 new comments

• Failed to add a new custom OpenAI Model via WEB UI

  #12767 commented on Jan 22, 2025 • 0 new comments

• Token Consumption Discrepancy in Dify LLM Node Due to Base64 Image Conversion

  #12631 commented on Jan 22, 2025 • 0 new comments

• [FeatureRequest] make the value CODE_MAX_STRING_LENGTH customizable at runtime for each workflow

  #9195 commented on Jan 22, 2025 • 0 new comments

• [FeatureRequest] Add a character counter and excessive text coloring in the input fields with a limit on number of characters

  #9149 commented on Jan 22, 2025 • 0 new comments

• [Feature Request] Display buttons for functions that does not have an equivalent input command on mobile browsers

  #9143 commented on Jan 22, 2025 • 0 new comments

• [BUG] The Command Palette and the contextual menu does not show in iOS Safari

  #9139 commented on Jan 22, 2025 • 0 new comments

• [FeatureRequest] Add a “Save to...” block, with the function to save the output files on SFTP or Clouds

  #9127 commented on Jan 22, 2025 • 0 new comments

• Error: Peer Closed Connection Without Sending Complete Message Body in Claude 3.5 (Sonnet 20240620)

  #12802 commented on Jan 22, 2025 • 0 new comments

• Regarding the issue of the default model value of the LLM node added through the node - adding function.

  #12797 commented on Jan 22, 2025 • 0 new comments

• Error in Tokens count for Iteration node

  #12801 commented on Jan 22, 2025 • 0 new comments

• no module named dify_plugin

  #12657 commented on Jan 22, 2025 • 0 new comments

• Endpoint for retrieving all conversations

  #12698 commented on Jan 21, 2025 • 0 new comments

• Add a time component for the selection of historical records.

  #11096 commented on Jan 21, 2025 • 0 new comments

• Support for Horizontal Scaling Across Components (API, Worker, Sandbox)

  #11902 commented on Jan 21, 2025 • 0 new comments

• Add re-ranking capabilities to web search results

  #11609 commented on Jan 21, 2025 • 0 new comments

• Markdown ingest is weird: text before first header is missing

  #11828 commented on Jan 21, 2025 • 0 new comments

• Can we add an optional feature to LLM nodes that only executes when a certain condition is met, otherwise it returns the default value

  #11726 commented on Jan 21, 2025 • 0 new comments

• app versions

  #10404 commented on Jan 21, 2025 • 0 new comments

• When I use the knowledge retrieval node and configure two external knowledge bases, the second configured knowledge base will definitely be invoked twice.

  #9500 commented on Jan 21, 2025 • 0 new comments

• only open suggested_questions

  #11579 commented on Jan 21, 2025 • 0 new comments

• After the Document indexing of the knowledge base fails, modify the segment settings, and save & process reports a "Document is not available" error. archive it and then unarchive it will report the "Document is being indexed, please try again later" error.

  #11746 commented on Jan 21, 2025 • 0 new comments

• Knowledge retrieval POST endpoints randomly returning 400 errors

  #11643 commented on Jan 21, 2025 • 0 new comments

• Incorporation of MarkITDown for enhanced

  #11737 commented on Jan 21, 2025 • 0 new comments

• It is recommended that the input fields of the start node can support sorting

  #10808 commented on Jan 21, 2025 • 0 new comments

• Migration files delete vector tables with pgvector in PostgreSQL

  #11835 commented on Jan 21, 2025 • 0 new comments

• Feature Request: Ability to Import Keywords Along with Questions and Answers in "Batch add chunks"

  #10810 commented on Jan 21, 2025 • 0 new comments

• Auth token unrecognized when logging in with email 2fa

  #11941 commented on Jan 21, 2025 • 0 new comments

• ChatFlow Node constant "RUNNING" status

  #11852 commented on Jan 21, 2025 • 0 new comments

• retrieve API cannot run in thread mode

  #11551 commented on Jan 21, 2025 • 0 new comments