I have searched for any existing and/or related issues.
I have searched for any existing and/or related discussions.
I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!).
I am using the latest version of Open WebUI.
Installation Method
Docker
Open WebUI Version
0.6.41
Ollama Version (if applicable)
No response
Operating System
Ubuntu 24.04
Browser (if applicable)
No response
Confirmation
I have read and followed all instructions in README.md.
I am using the latest version of both Open WebUI and Ollama.
I have included the browser console logs.
I have included the Docker container logs.
I have provided every relevant configuration, setting, and environment variable used in my setup.
I have clearly listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
I have documented step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation. My steps:
Start with the initial platform/version/OS and dependencies used,
Specify exact install/launch/configure commands,
List URLs visited, user input (incl. example values/emails/passwords if needed),
Describe all options and toggles enabled or changed,
Include any files or environmental changes,
Identify the expected and actual result at each stage,
Ensure any reasonably skilled user can follow and hit the same issue.
Expected Behavior
The response should render a mermaid diagram.
Actual Behavior
The response says
"Failed to render diagram: Parse error on line 4:
...Array Sorted 0s | 1s | 2s])
-----------------------^
Expecting 'STADIUMEND', 'TAGEND', 'UNICODE_TEXT', 'TEXT', 'TAGSTART', got 'PIPE'
Steps to Reproduce
Launch OpenWebUI workspace.
Send request: "Create a mermaid diagram that shows the Dutch National Flag algorithm."
Look at response.
Logs & Screenshots
logs:
[{'role': 'user', 'content': '### Task:\nGenerate 1-3 broad tags categorizing the main themes of the chat history, along with 1-3 more specific subtopic tags.\n\n### Guidelines:\n- Start with high-level domains (e.g. Science, Technology, Philosophy, Arts, Politics, Business, Health, Sports, Entertainment, Education)\n- Consider including relevant subfields/subdomains if they are strongly represented throughout the conversation\n- If content is too short (less than 3 messages) or too diverse, use only ["General"]\n- Use the chat's primary language; default to English if multilingual\n- Prioritize accuracy over specificity\n\n### Output:\nJSON format: { "tags": ["tag1", "tag2", "tag3"] }\n\n### Chat History:\n<chat_history>\nUSER: create a mermaid diagram that shows the Dutch National Flag algorithm.\nASSISTANT: mermaid\nflowchart TD\n Start([Start: Unsorted Array]) --> Init[Initialize three pointers:<br/>low = 0, mid = 0, high = n-1]\n Init --> Check{mid <= high?}\n \n Check -->|No| End([End: Array Sorted<br/>0s | 1s | 2s])\n \n Check -->|Yes| Examine{Examine<br/>arr[mid]}\n \n Examine -->|0| Case0[Swap arr[low] with arr[mid]<br/>low++<br/>mid++]\n Examine -->|1| Case1[mid++]\n Examine -->|2| Case2[Swap arr[mid] with arr[high]<br/>high--]\n \n Case0 --> Check\n Case1 --> Check\n Case2 --> Check\n \n style Start fill:#e1f5ff\n style End fill:#d4edda\n style Examine fill:#fff3cd\n style Case0 fill:#f8d7da\n style Case1 fill:#d1ecf1\n style Case2 fill:#f8d7da\n\n\nThis diagram illustrates the Dutch National Flag algorithm (created by Edsger Dijkstra) which sorts an array containing three distinct values (typically 0, 1, and 2) in a single pass.\n\nKey concepts:\n- low pointer: marks the boundary of 0s (everything before low is 0)\n- mid pointer: current element being examined\n- high pointer: marks the boundary of 2s (everything after high is 2)\n\nThe algorithm partitions the array into three sections:\n1. [0...low-1] → all 0s\n2. [low...mid-1] → all 1s\n3. [mid...high] → unknown/unprocessed\n4. [high+1...n-1] → all 2s\n</chat_history>'}]
Additional Information
This was not occurring in 0.6.40. With the update to the underlying data model, we cannot revert back to previous versions.
I am using Claude. This occurs in multipe Claude models: Claude 4.5 Sonnet, Claude 4 Sonnet, Claude 3.7 Sonnet, and Claude 3.5 Sonnet v2
Originally created by @j63440490 on GitHub (Dec 17, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/20019
### Check Existing Issues
- [x] I have searched for any existing and/or related issues.
- [x] I have searched for any existing and/or related discussions.
- [x] I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!).
- [x] I am using the latest version of Open WebUI.
### Installation Method
Docker
### Open WebUI Version
0.6.41
### Ollama Version (if applicable)
_No response_
### Operating System
Ubuntu 24.04
### Browser (if applicable)
_No response_
### Confirmation
- [x] I have read and followed all instructions in `README.md`.
- [x] I am using the latest version of **both** Open WebUI and Ollama.
- [x] I have included the browser console logs.
- [x] I have included the Docker container logs.
- [x] I have **provided every relevant configuration, setting, and environment variable used in my setup.**
- [x] I have clearly **listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup** (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
- [x] I have documented **step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation**. My steps:
- Start with the initial platform/version/OS and dependencies used,
- Specify exact install/launch/configure commands,
- List URLs visited, user input (incl. example values/emails/passwords if needed),
- Describe all options and toggles enabled or changed,
- Include any files or environmental changes,
- Identify the expected and actual result at each stage,
- Ensure any reasonably skilled user can follow and hit the same issue.
### Expected Behavior
The response should render a mermaid diagram.
### Actual Behavior
The response says
"Failed to render diagram: Parse error on line 4:
...Array Sorted<br/>0s | 1s | 2s])
-----------------------^
Expecting 'STADIUMEND', 'TAGEND', 'UNICODE_TEXT', 'TEXT', 'TAGSTART', got 'PIPE'
### Steps to Reproduce
1. Launch OpenWebUI workspace.
2. Send request: "Create a mermaid diagram that shows the Dutch National Flag algorithm."
3. Look at response.
### Logs & Screenshots
<img width="1279" height="805" alt="Image" src="https://github.com/user-attachments/assets/e8548b1a-8e41-492a-ae27-305eba186c70" />
logs:
[{'role': 'user', 'content': '### Task:\nGenerate 1-3 broad tags categorizing the main themes of the chat history, along with 1-3 more specific subtopic tags.\n\n### Guidelines:\n- Start with high-level domains (e.g. Science, Technology, Philosophy, Arts, Politics, Business, Health, Sports, Entertainment, Education)\n- Consider including relevant subfields/subdomains if they are strongly represented throughout the conversation\n- If content is too short (less than 3 messages) or too diverse, use only ["General"]\n- Use the chat\'s primary language; default to English if multilingual\n- Prioritize accuracy over specificity\n\n### Output:\nJSON format: { "tags": ["tag1", "tag2", "tag3"] }\n\n### Chat History:\n<chat_history>\nUSER: create a mermaid diagram that shows the Dutch National Flag algorithm.\nASSISTANT: ```mermaid\nflowchart TD\n Start([Start: Unsorted Array]) --> Init[Initialize three pointers:<br/>low = 0, mid = 0, high = n-1]\n Init --> Check{mid <= high?}\n \n Check -->|No| End([End: Array Sorted<br/>0s | 1s | 2s])\n \n Check -->|Yes| Examine{Examine<br/>arr[mid]}\n \n Examine -->|0| Case0[Swap arr[low] with arr[mid]<br/>low++<br/>mid++]\n Examine -->|1| Case1[mid++]\n Examine -->|2| Case2[Swap arr[mid] with arr[high]<br/>high--]\n \n Case0 --> Check\n Case1 --> Check\n Case2 --> Check\n \n style Start fill:#e1f5ff\n style End fill:#d4edda\n style Examine fill:#fff3cd\n style Case0 fill:#f8d7da\n style Case1 fill:#d1ecf1\n style Case2 fill:#f8d7da\n```\n\nThis diagram illustrates the **Dutch National Flag algorithm** (created by Edsger Dijkstra) which sorts an array containing three distinct values (typically 0, 1, and 2) in a single pass.\n\n**Key concepts:**\n- **low pointer**: marks the boundary of 0s (everything before low is 0)\n- **mid pointer**: current element being examined\n- **high pointer**: marks the boundary of 2s (everything after high is 2)\n\n**The algorithm partitions the array into three sections:**\n1. `[0...low-1]` → all 0s\n2. `[low...mid-1]` → all 1s\n3. `[mid...high]` → unknown/unprocessed\n4. `[high+1...n-1]` → all 2s\n</chat_history>'}]
### Additional Information
This was not occurring in 0.6.40. With the update to the underlying data model, we cannot revert back to previous versions.
I am using Claude. This occurs in multipe Claude models: Claude 4.5 Sonnet, Claude 4 Sonnet, Claude 3.7 Sonnet, and Claude 3.5 Sonnet v2
GiteaMirror
added the bug label 2026-04-25 08:38:48 -05:00
If this is a duplicate, please consider closing this issue and adding any additional details to the existing one
If you found a solution in any of these issues, please share it here to help others
This comment was generated automatically by a bot. Please react with a 👍 if this comment was helpful, or a 👎 if it was not.
<!-- gh-comment-id:3666997694 -->
@owui-terminator[bot] commented on GitHub (Dec 17, 2025):
🔍 **Similar Issues Found**
I found some existing issues that might be related to this one. Please check if any of these are duplicates or contain helpful solutions:
1. [#19777](https://github.com/open-webui/open-webui/issues/19777) **issue:**
*by Yaute7 • Dec 05, 2025 • `bug`*
2. [#19864](https://github.com/open-webui/open-webui/issues/19864) **issue:**
*by Haervwe • Dec 10, 2025 • `bug`*
3. [#19877](https://github.com/open-webui/open-webui/issues/19877) **issue:**
*by dotmobo • Dec 11, 2025 • `bug`*
4. [#19861](https://github.com/open-webui/open-webui/issues/19861) **issue:**
*by QuitHub • Dec 10, 2025 • `bug`*
5. [#19563](https://github.com/open-webui/open-webui/issues/19563) **issue:**
*by naruto7g • Nov 28, 2025 • `bug`*
<details>
<summary>Show 5 more related issues</summary>
6. [#19211](https://github.com/open-webui/open-webui/issues/19211) **issue:**
*by Byrnes9 • Nov 16, 2025 • `bug`*
7. [#16959](https://github.com/open-webui/open-webui/issues/16959) **issue:**
*by Te-eMster • Aug 27, 2025 • `bug`*
8. [#15638](https://github.com/open-webui/open-webui/issues/15638) **issue:**
*by PaulX1029 • Jul 11, 2025 • `bug`*
9. [#17391](https://github.com/open-webui/open-webui/issues/17391) **issue:**
*by abxis • Sep 12, 2025 • `bug`*
10. [#17388](https://github.com/open-webui/open-webui/issues/17388) **issue:**
*by abxis • Sep 12, 2025 • `bug`*
</details>
---
💡 **Tips:**
- If this is a duplicate, please consider closing this issue and adding any additional details to the existing one
- If you found a solution in any of these issues, please share it here to help others
*This comment was generated automatically by a bot.* Please react with a 👍 if this comment was helpful, or a 👎 if it was not.
@pr-validator-bot commented on GitHub (Dec 17, 2025):
⚠️ Invalid Issue Title
Please provide a descriptive title for your issue. Titles that are empty, very short (under 10 characters), or generic (like "issue:") make it difficult for volunteer contributors to understand and triage issues.
Please update the title to reflect the content of your issue.
<!-- gh-comment-id:3667024464 -->
@pr-validator-bot commented on GitHub (Dec 17, 2025):
# ⚠️ Invalid Issue Title
Please provide a descriptive title for your issue. Titles that are empty, very short (under 10 characters), or generic (like "issue:") make it difficult for volunteer contributors to understand and triage issues.
Please update the title to reflect the content of your issue.
Can you prove that the same chat was rendered differently in v0.6.41 compared to v0.6.40? I recommend cloning the chat on v0.6.40 so that you have a "backup" chat to test this with. Simply put, this looks like an error with the model failing to provide valid Mermaid Diagram code to be rendered.
<!-- gh-comment-id:3667433386 -->
@silentoplayz commented on GitHub (Dec 17, 2025):
Can you prove that the same chat was rendered differently in v0.6.41 compared to v0.6.40? I recommend cloning the chat on v0.6.40 so that you have a "backup" chat to test this with. Simply put, this looks like an error with the model failing to provide valid Mermaid Diagram code to be rendered.
I tested this on v0.6.41 and v0.6.40, in both versions the Models always fail to generate a mermaid from the prompt "Create a mermaid diagram that shows the Dutch National Flag algorithm". The error seems to always be unquoted node/decision text.
Outputted Mermaid:
flowchart TD
A[Start] --> B[Initialize<br/>low = 0<br/>mid = 0<br/>high = n - 1]
B --> C{mid <= high}
C -- "No" --> Z[End\nArray is sorted]
C -- "Yes" --> D{arr[mid]}
D -- "0" --> E[Swap arr[low] & arr[mid]]
E --> F[low++\nmid++]
F --> C
D -- "1" --> G[mid++]
G --> C
D -- "2" --> H[Swap arr[mid] & arr[high]]
H --> I[high--]
I --> C
Github mermaid render shows the same error:
flowchart TD
A[Start] --> B[Initialize<br/>low = 0<br/>mid = 0<br/>high = n - 1]
B --> C{mid <= high}
C -- "No" --> Z[End\nArray is sorted]
C -- "Yes" --> D{arr[mid]}
D -- "0" --> E[Swap arr[low] & arr[mid]]
E --> F[low++\nmid++]
F --> C
D -- "1" --> G[mid++]
G --> C
D -- "2" --> H[Swap arr[mid] & arr[high]]
H --> I[high--]
I --> C
Asking the model to "wrap all labels in quotes" gets it to consistently output a mermaid diagram without errors.
Fixed Mermaid:
flowchart TD
A["Start"] --> B["Initialize<br/>low = 0<br/>mid = 0<br/>high = n - 1"]
B --> C{"mid <= high"}
C -- "No" --> Z["End<br/>Array is sorted"]
C -- "Yes" --> D{"arr[mid]"}
D -- "0" --> E["Swap arr[low] and arr[mid]"]
E --> F["low++<br/>mid++"]
F --> C
D -- "1" --> G["mid++"]
G --> C
D -- "2" --> H["Swap arr[mid] and arr[high]"]
H --> I["high--"]
I --> C
Seems to be a issue on the models side, as silentoplayz said. :)
<!-- gh-comment-id:3668706678 -->
@rozatoo commented on GitHub (Dec 18, 2025):
I tested this on v0.6.41 and v0.6.40, in both versions the Models always fail to generate a mermaid from the prompt "Create a mermaid diagram that shows the Dutch National Flag algorithm". The error seems to always be unquoted node/decision text.
Outputted Mermaid:
```text
flowchart TD
A[Start] --> B[Initialize<br/>low = 0<br/>mid = 0<br/>high = n - 1]
B --> C{mid <= high}
C -- "No" --> Z[End\nArray is sorted]
C -- "Yes" --> D{arr[mid]}
D -- "0" --> E[Swap arr[low] & arr[mid]]
E --> F[low++\nmid++]
F --> C
D -- "1" --> G[mid++]
G --> C
D -- "2" --> H[Swap arr[mid] & arr[high]]
H --> I[high--]
I --> C
```
Github mermaid render shows the same error:
```mermaid
flowchart TD
A[Start] --> B[Initialize<br/>low = 0<br/>mid = 0<br/>high = n - 1]
B --> C{mid <= high}
C -- "No" --> Z[End\nArray is sorted]
C -- "Yes" --> D{arr[mid]}
D -- "0" --> E[Swap arr[low] & arr[mid]]
E --> F[low++\nmid++]
F --> C
D -- "1" --> G[mid++]
G --> C
D -- "2" --> H[Swap arr[mid] & arr[high]]
H --> I[high--]
I --> C
```
Asking the model to "wrap all labels in quotes" gets it to consistently output a mermaid diagram without errors.
Fixed Mermaid:
```text
flowchart TD
A["Start"] --> B["Initialize<br/>low = 0<br/>mid = 0<br/>high = n - 1"]
B --> C{"mid <= high"}
C -- "No" --> Z["End<br/>Array is sorted"]
C -- "Yes" --> D{"arr[mid]"}
D -- "0" --> E["Swap arr[low] and arr[mid]"]
E --> F["low++<br/>mid++"]
F --> C
D -- "1" --> G["mid++"]
G --> C
D -- "2" --> H["Swap arr[mid] and arr[high]"]
H --> I["high--"]
I --> C
```
Seems to be a issue on the models side, as silentoplayz said. :)
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @j63440490 on GitHub (Dec 17, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/20019
Check Existing Issues
Installation Method
Docker
Open WebUI Version
0.6.41
Ollama Version (if applicable)
No response
Operating System
Ubuntu 24.04
Browser (if applicable)
No response
Confirmation
README.md.Expected Behavior
The response should render a mermaid diagram.
Actual Behavior
The response says
"Failed to render diagram: Parse error on line 4:
...Array Sorted
0s | 1s | 2s])
-----------------------^
Expecting 'STADIUMEND', 'TAGEND', 'UNICODE_TEXT', 'TEXT', 'TAGSTART', got 'PIPE'
Steps to Reproduce
Logs & Screenshots
logs:
[{'role': 'user', 'content': '### Task:\nGenerate 1-3 broad tags categorizing the main themes of the chat history, along with 1-3 more specific subtopic tags.\n\n### Guidelines:\n- Start with high-level domains (e.g. Science, Technology, Philosophy, Arts, Politics, Business, Health, Sports, Entertainment, Education)\n- Consider including relevant subfields/subdomains if they are strongly represented throughout the conversation\n- If content is too short (less than 3 messages) or too diverse, use only ["General"]\n- Use the chat's primary language; default to English if multilingual\n- Prioritize accuracy over specificity\n\n### Output:\nJSON format: { "tags": ["tag1", "tag2", "tag3"] }\n\n### Chat History:\n<chat_history>\nUSER: create a mermaid diagram that shows the Dutch National Flag algorithm.\nASSISTANT:
mermaid\nflowchart TD\n Start([Start: Unsorted Array]) --> Init[Initialize three pointers:<br/>low = 0, mid = 0, high = n-1]\n Init --> Check{mid <= high?}\n \n Check -->|No| End([End: Array Sorted<br/>0s | 1s | 2s])\n \n Check -->|Yes| Examine{Examine<br/>arr[mid]}\n \n Examine -->|0| Case0[Swap arr[low] with arr[mid]<br/>low++<br/>mid++]\n Examine -->|1| Case1[mid++]\n Examine -->|2| Case2[Swap arr[mid] with arr[high]<br/>high--]\n \n Case0 --> Check\n Case1 --> Check\n Case2 --> Check\n \n style Start fill:#e1f5ff\n style End fill:#d4edda\n style Examine fill:#fff3cd\n style Case0 fill:#f8d7da\n style Case1 fill:#d1ecf1\n style Case2 fill:#f8d7da\n\n\nThis diagram illustrates the Dutch National Flag algorithm (created by Edsger Dijkstra) which sorts an array containing three distinct values (typically 0, 1, and 2) in a single pass.\n\nKey concepts:\n- low pointer: marks the boundary of 0s (everything before low is 0)\n- mid pointer: current element being examined\n- high pointer: marks the boundary of 2s (everything after high is 2)\n\nThe algorithm partitions the array into three sections:\n1.[0...low-1]→ all 0s\n2.[low...mid-1]→ all 1s\n3.[mid...high]→ unknown/unprocessed\n4.[high+1...n-1]→ all 2s\n</chat_history>'}]Additional Information
This was not occurring in 0.6.40. With the update to the underlying data model, we cannot revert back to previous versions.
I am using Claude. This occurs in multipe Claude models: Claude 4.5 Sonnet, Claude 4 Sonnet, Claude 3.7 Sonnet, and Claude 3.5 Sonnet v2
@owui-terminator[bot] commented on GitHub (Dec 17, 2025):
🔍 Similar Issues Found
I found some existing issues that might be related to this one. Please check if any of these are duplicates or contain helpful solutions:
#19777 issue:
by Yaute7 • Dec 05, 2025 •
bug#19864 issue:
by Haervwe • Dec 10, 2025 •
bug#19877 issue:
by dotmobo • Dec 11, 2025 •
bug#19861 issue:
by QuitHub • Dec 10, 2025 •
bug#19563 issue:
by naruto7g • Nov 28, 2025 •
bugShow 5 more related issues
#19211 issue:
by Byrnes9 • Nov 16, 2025 •
bug#16959 issue:
by Te-eMster • Aug 27, 2025 •
bug#15638 issue:
by PaulX1029 • Jul 11, 2025 •
bug#17391 issue:
by abxis • Sep 12, 2025 •
bug#17388 issue:
by abxis • Sep 12, 2025 •
bug💡 Tips:
This comment was generated automatically by a bot. Please react with a 👍 if this comment was helpful, or a 👎 if it was not.
@pr-validator-bot commented on GitHub (Dec 17, 2025):
⚠️ Invalid Issue Title
Please provide a descriptive title for your issue. Titles that are empty, very short (under 10 characters), or generic (like "issue:") make it difficult for volunteer contributors to understand and triage issues.
Please update the title to reflect the content of your issue.
@silentoplayz commented on GitHub (Dec 17, 2025):
Can you prove that the same chat was rendered differently in v0.6.41 compared to v0.6.40? I recommend cloning the chat on v0.6.40 so that you have a "backup" chat to test this with. Simply put, this looks like an error with the model failing to provide valid Mermaid Diagram code to be rendered.
@rozatoo commented on GitHub (Dec 18, 2025):
I tested this on v0.6.41 and v0.6.40, in both versions the Models always fail to generate a mermaid from the prompt "Create a mermaid diagram that shows the Dutch National Flag algorithm". The error seems to always be unquoted node/decision text.
Outputted Mermaid:
Github mermaid render shows the same error:
Asking the model to "wrap all labels in quotes" gets it to consistently output a mermaid diagram without errors.
Fixed Mermaid:
Seems to be a issue on the models side, as silentoplayz said. :)