### What problem does this PR solve?
Add tree_merge for law parsers, significantly outperforming
hierarchical_merge, solved: #8637
1. Add tree_merge for law parsers, include build_tree and get_tree by
dfs.
2. add Copyright statement for helath_utils
### Type of change
- [x] Documentation Update
- [x] Performance Improvement
### What problem does this PR solve?
Add a chat widget. I'll probably need some assistance to get this ready
for merge!
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
Co-authored-by: Mohamed Mathari <nocodeventure@Mac-mini-van-Mohamed.fritz.box>
### What problem does this PR solve?
Introduce new feature: RAGFlow system admin service and CLI
### Introduction
Admin Service is a dedicated management component designed to monitor,
maintain, and administrate the RAGFlow system. It provides comprehensive
tools for ensuring system stability, performing operational tasks, and
managing users and permissions efficiently.
The service offers monitoring of critical components, including the
RAGFlow server, Task Executor processes, and dependent services such as
MySQL, Infinity / Elasticsearch, Redis, and MinIO. It automatically
checks their health status, resource usage, and uptime, and performs
restarts in case of failures to minimize downtime.
For user and system management, it supports listing, creating,
modifying, and deleting users and their associated resources like
knowledge bases and Agents.
Built with scalability and reliability in mind, the Admin Service
ensures smooth system operation and simplifies maintenance workflows.
It consists of a server-side Service and a command-line client (CLI),
both implemented in Python. User commands are parsed using the Lark
parsing toolkit.
- **Admin Service**: A backend service that interfaces with the RAGFlow
system to execute administrative operations and monitor its status.
- **Admin CLI**: A command-line interface that allows users to connect
to the Admin Service and issue commands for system management.
### Starting the Admin Service
1. Before start Admin Service, please make sure RAGFlow system is
already started.
2. Run the service script:
```bash
python admin/admin_server.py
```
The service will start and listen for incoming connections from the CLI
on the configured port.
### Using the Admin CLI
1. Ensure the Admin Service is running.
2. Launch the CLI client:
```bash
python admin/admin_client.py -h 0.0.0.0 -p 9381
## Supported Commands
Commands are case-insensitive and must be terminated with a semicolon
(`;`).
### Service Management Commands
- [x] `LIST SERVICES;`
- Lists all available services within the RAGFlow system.
- [ ] `SHOW SERVICE <id>;`
- Shows detailed status information for the service identified by
`<id>`.
- [ ] `STARTUP SERVICE <id>;`
- Attempts to start the service identified by `<id>`.
- [ ] `SHUTDOWN SERVICE <id>;`
- Attempts to gracefully shut down the service identified by `<id>`.
- [ ] `RESTART SERVICE <id>;`
- Attempts to restart the service identified by `<id>`.
### User Management Commands
- [x] `LIST USERS;`
- Lists all users known to the system.
- [ ] `SHOW USER '<username>';`
- Shows details and permissions for the specified user. The username
must be enclosed in single or double quotes.
- [ ] `DROP USER '<username>';`
- Removes the specified user from the system. Use with caution.
- [ ] `ALTER USER PASSWORD '<username>' '<new_password>';`
- Changes the password for the specified user.
### Data and Agent Commands
- [ ] `LIST DATASETS OF '<username>';`
- Lists the datasets associated with the specified user.
- [ ] `LIST AGENTS OF '<username>';`
- Lists the agents associated with the specified user.
### Meta-Commands
Meta-commands are prefixed with a backslash (`\`).
- `\?` or `\help`
- Shows help information for the available commands.
- `\q` or `\quit`
- Exits the CLI application.
## Examples
```commandline
admin> list users;
+-------------------------------+------------------------+-----------+-------------+
| create_date | email | is_active | nickname |
+-------------------------------+------------------------+-----------+-------------+
| Fri, 22 Nov 2024 16:03:41 GMT | jeffery@infiniflow.org | 1 | Jeffery |
| Fri, 22 Nov 2024 16:10:55 GMT | aya@infiniflow.org | 1 | Waterdancer |
+-------------------------------+------------------------+-----------+-------------+
admin> list services;
+-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+
| extra | host | id | name | port | service_type |
+-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+
| {} | 0.0.0.0 | 0 | ragflow_0 | 9380 | ragflow_server |
| {'meta_type': 'mysql', 'password': 'infini_rag_flow', 'username': 'root'} | localhost | 1 | mysql | 5455 | meta_data |
| {'password': 'infini_rag_flow', 'store_type': 'minio', 'user': 'rag_flow'} | localhost | 2 | minio | 9000 | file_store |
| {'password': 'infini_rag_flow', 'retrieval_type': 'elasticsearch', 'username': 'elastic'} | localhost | 3 | elasticsearch | 1200 | retrieval |
| {'db_name': 'default_db', 'retrieval_type': 'infinity'} | localhost | 4 | infinity | 23817 | retrieval |
| {'database': 1, 'mq_type': 'redis', 'password': 'infini_rag_flow'} | localhost | 5 | redis | 6379 | message_queue |
+-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+
```
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
Signed-off-by: jinhai <haijin.chn@gmail.com>
### What problem does this PR solve?
This PR is related to
[#9961](https://github.com/infiniflow/ragflow/issues/9961).
In the Chat Settings screen, the textarea did not support scrolling when
the content grew longer than its visible area, which made it less
convenient to use.
Also, there was no Japanese placeholder text to guide users on what to
enter in the field.
This PR improves the user experience by:
- Adding `overflow-y-auto` to the textarea so that long content can be
scrolled smoothly.
- Introducing a placeholder (`メッセージを入力してください...`) to provide clearer
guidance for users.
https://github.com/user-attachments/assets/95553331-087b-42c5-a41d-5dfe08047bae
### What has been considered
As an alternative solution, I explored replacing the textarea with the
existing `PromptEditor` component.
However, this approach triggered a `canvas not found.` alert.
The current implementation of `PromptEditor` internally attempts to
fetch **agent (canvas) information**, but in the Chat Settings screen no
such ID exists. As a result, the API call fails and the backend returns
`canvas not found.`.
One possible workaround would be to extend `PromptEditor` with a
**“disable variable picker” flag**, ensuring that plugins are not loaded
in contexts like Chat Settings. While feasible, this would have a
broader impact across the codebase.
Given these considerations, I decided to address the issue in a simpler
way by applying a Tailwind utility (`overflow-y-auto`). Since the UI
design is expected to change in the future, this solution is considered
sufficient for now.
<img width="1501" height="794" alt="Screenshot 2025-09-20 at 15 00 12"
src="https://github.com/user-attachments/assets/85578ee8-489f-4ede-b3af-bafd7afe95bd"
/>
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
- [ ] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):
### What problem does this PR solve?
Skip `tag_query` step if `tag_kbs` are empty.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
issue:
[Bug]: anthropic model have not baseurl selecting,need add #8546
change:
This PR adds support for using Anthropic models through a third-party
API by allowing a custom base_url.
It ensures compatibility with both the official Anthropic endpoint and
external providers.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
fix the rerank_model condition logic by correcting the np.isclose check.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Support server health check. Solved issue: #10106
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Add file convert to document API just like file2document_app.py
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Fix: Fixed the knowledge base's embedded model form layout and
dependency imports in the main branch.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Migrate OpenAI-compatible chats to LiteLLM.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Fix: Merge different types of models from the same manufacturer #10146
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Handle zero and nan in calculate.
#10125
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Related PR:
Feat: add CometAPI to LLMFactory and update related mappings #10119
Change:
Fixes the issue where the embedding model in CometAPI was not being
called correctly
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)
---------
Co-authored-by: TensorNull <tensor.null@gmail.com>
### What problem does this PR solve?
Add support for KB document basic info
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### Related issues
#10078
### What problem does this PR solve?
Integrate CometAPI provider.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
- [x] Documentation Update
### What problem does this PR solve?
The original text for vectorSimilarityWeight in Chinese version was
"相似度相似度权重," which is obviously a malformed phrase. It has now been
changed to "向量相似度权重". Also, align it with the English version 'Vector
similarity weight'.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Support image recognition with image links in markdown files, solved
issue: #8755
Fixed log info error in code_exec, solved issue: #10064
### Type of change (8755)
- [x] New Feature (non-breaking change which adds functionality)
### Type of change (10064)
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Dataflow support audio. And fix giteeAI's sequence2text model.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
An exception happens if you give session_id to agent_open_ai completion.
Because session_id is being given as well as **req so it tries to send
session_id twice. But also the logic seemed odd on picking one of
session_id, id, metadata.id. So cleaned it up a little.
See #10111
### Type of change
- [X] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
This PR fixes incorrect naming for PostgreSQL usage by replacing all
instances of `postgresql` with the correct `postgres` in the `db_type`
field. This resolves potential configuration errors and ensures
consistency when specifying the database type.
Also fixed handling of None for `get_queue_length`
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
Co-authored-by: cucusenok <BP-116: updated readme.md>
### What problem does this PR solve?
issue:
[Bug]: Agent component (HTTP Request) "'>' not supported between
instances of 'int' and 'NoneType'"
[#10096](https://github.com/infiniflow/ragflow/issues/10096)
Change:
When the Invoke class instantiates HtmlParser without providing the
chunk_token_num parameter, the value defaults to None, leading to a
comparison error with block_token_count.
This change sets the default chunk_token_num to 512 to prevent such
errors.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
Co-authored-by: BadwomanCraZY <511528396@qq.com>
### What problem does this PR solve?
support parse image by OCR or VLM.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Add support for the Ascend table structure recognizer.
Use the environment variable `TABLE_STRUCTURE_RECOGNIZER_TYPE=ascend` to
enable the Ascend table structure recognizer.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Updated SQL assistant template to wrap variables like 'sys.query' and
'Agent:WickedGoatsDivide@content' in curly braces for better template
variable syntax consistency.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Supports Ascend layout recognizer.
Use the environment variable `LAYOUT_RECOGNIZER_TYPE=ascend` to enable
the Ascend layout recognizer, and `ASCEND_LAYOUT_RECOGNIZER_DEVICE_ID=n`
(for example, n=0) to specify the Ascend device ID.
Ensure that you have installed the [ais
tools](https://gitee.com/ascend/tools/tree/master/ais-bench_workload/tool/ais_bench)
properly.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
judge not empty before delete session.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Fix: The same model appears twice in the drop-down box. #10102
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
…release workflow (#10039)
This change updates the GitHub Actions workflow to push additional
stable tags alongside version tags, enabling automated update tools like
Watchtower to detect and pull the latest images correctly.
Refs:
[https://github.com/infiniflow/ragflow/issues/10039](https://github.com/infiniflow/ragflow/issues/10039)
### What problem does this PR solve?
Automated container update tools such as Watchtower rely on stable tags
like `latest` to identify the newest images. Previously, only
version-specific tags were pushed, which prevented these tools from
detecting new releases automatically. This PR adds multiple stable tags
(`latest-full`, `latest-slim`) alongside version tags to the Docker
image publishing workflow, ensuring smooth and reliable automated
updates without manual tag management.
### Type of change
- [ ] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):
---------
Co-authored-by: Zhichang Yu <yuzhichang@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
### What problem does this PR solve?
terminate onnx inference session and release memory manually.
Issue #5050
Issue #9992
Issue #8805
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Feat: Translate the fields of the embedded dialog box on the agent page
#3221
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
The chat dialog box cannot be fully displayed on a small screen #10034
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Fix: Hide dataflow related functions #9869
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
fix text input exceed token num limit when using siliconflow's embedding
model BAAI/bge-large-zh-v1.5 and BAAI/bge-large-en-v1.5, truncate before
input.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
