I have searched the existing issues and discussions.
I am using the latest version of Open WebUI.
Installation Method
Other
Open WebUI Version
v0.6.10
Ollama Version (if applicable)
No response
Operating System
Debian 12, Docker Compose
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 listed steps to reproduce the bug in detail.
Expected Behavior
Querying an LLM using native functions with a large OpenAPI tool equipped should provide some way of filtering or disabling available endpoints down to a subset of functionality, to remain within LLM limits.
I have observed this issue on two separate OpenWebUI instances, both with external OpenAPI endpoints and also via MCPO.
Actual Behavior
Adding the OpenWebUI openapi.json as a system-wide tool, and querying GPT-4.1 with that tool equipped, produces an error 400: Invalid 'tools': array too long. Expected an array with maximum length 128, but got an array with length 315 instead.
Steps to Reproduce
In Admin Panel -> Tools, add a tool server for https:///openapi.json. Use 'Verify Connection' to ensure it's connecting correctly.
Refresh the browser, then open a new chat with OpenAI's gpt-4.1 or o4-mini.
In Chat Controls, set Function Calling to 'native'.
In the message bar, click the + and enable the 'Open WebUI' endpoint.
Send any message to the LLM.
Error message as received: "400: Invalid 'tools': array too long. Expected an array with maximum length 128, but got an array with length 315 instead."
Originally created by @ciejer on GitHub (May 26, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/14360
### Check Existing Issues
- [x] I have searched the existing issues and discussions.
- [x] I am using the latest version of Open WebUI.
### Installation Method
Other
### Open WebUI Version
v0.6.10
### Ollama Version (if applicable)
_No response_
### Operating System
Debian 12, Docker Compose
### 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 listed steps to reproduce the bug in detail.
### Expected Behavior
Querying an LLM using native functions with a large OpenAPI tool equipped should provide some way of filtering or disabling available endpoints down to a subset of functionality, to remain within LLM limits.
I have observed this issue on two separate OpenWebUI instances, both with external OpenAPI endpoints and also via MCPO.
### Actual Behavior
Adding the OpenWebUI openapi.json as a system-wide tool, and querying GPT-4.1 with that tool equipped, produces an error 400: Invalid 'tools': array too long. Expected an array with maximum length 128, but got an array with length 315 instead.
### Steps to Reproduce
1. In Admin Panel -> Tools, add a tool server for https://<your openwebui url>/openapi.json. Use 'Verify Connection' to ensure it's connecting correctly.
2. Refresh the browser, then open a new chat with OpenAI's gpt-4.1 or o4-mini.
3. In Chat Controls, set Function Calling to 'native'.
4. In the message bar, click the + and enable the 'Open WebUI' endpoint.
5. Send any message to the LLM.
6. Error message as received: "400: Invalid 'tools': array too long. Expected an array with maximum length 128, but got an array with length 315 instead."
### Logs & Screenshots
Browser Console Logs:
[openwebui.dataops.nz-1748294240937.log](https://github.com/user-attachments/files/20445996/openwebui.dataops.nz-1748294240937.log)
Docker Logs:
[docker_logs.txt](https://github.com/user-attachments/files/20445997/docker_logs.txt)
### Additional Information
_No response_
GiteaMirror
added the bug label 2026-04-25 06:37:02 -05:00
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 @ciejer on GitHub (May 26, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/14360
Check Existing Issues
Installation Method
Other
Open WebUI Version
v0.6.10
Ollama Version (if applicable)
No response
Operating System
Debian 12, Docker Compose
Browser (if applicable)
No response
Confirmation
README.md.Expected Behavior
Querying an LLM using native functions with a large OpenAPI tool equipped should provide some way of filtering or disabling available endpoints down to a subset of functionality, to remain within LLM limits.
I have observed this issue on two separate OpenWebUI instances, both with external OpenAPI endpoints and also via MCPO.
Actual Behavior
Adding the OpenWebUI openapi.json as a system-wide tool, and querying GPT-4.1 with that tool equipped, produces an error 400: Invalid 'tools': array too long. Expected an array with maximum length 128, but got an array with length 315 instead.
Steps to Reproduce
Logs & Screenshots
Browser Console Logs:
openwebui.dataops.nz-1748294240937.log
Docker Logs:
docker_logs.txt
Additional Information
No response
@tjbck commented on GitHub (May 26, 2025):
OpenAI needs to support more tools to be included in the chat request.