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
0.9.5
Ollama Version (if applicable)
Operating System
Browser (if applicable)
Chrome, but also tested on Safari
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
Actual Behavior
I always had a single ollama client connected to my OWUI instance, but since i've also added vLLM, I get:
Uncaught Error: https://svelte.dev/e/each_key_duplicate
at ht (Selector.svelte:734:13)
Every time I try to open the dropdown menu for model selection
Steps to Reproduce
I think the problem is if the model name of vLLM is also present in Ollama, then they clash and break the whole model selectio menu.
So I guess that having an Ollama connection with model X, and a vLLM that exposes a model with the same X name will crash it correctly
Logs & Screenshots
Additional Information
No response
Originally created by @AlbertoSinigaglia on GitHub (May 11, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/24578
### 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
0.9.5
### Ollama Version (if applicable)
-
### Operating System
-
### Browser (if applicable)
Chrome, but also tested on Safari
### 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
-
### Actual Behavior
I always had a single ollama client connected to my OWUI instance, but since i've also added vLLM, I get:
```
Uncaught Error: https://svelte.dev/e/each_key_duplicate
at ht (Selector.svelte:734:13)
```
Every time I try to open the dropdown menu for model selection
### Steps to Reproduce
I think the problem is if the model name of vLLM is also present in Ollama, then they clash and break the whole model selectio menu.
So I guess that having an Ollama connection with model X, and a vLLM that exposes a model with the same X name will crash it correctly
### Logs & Screenshots
<img width="1674" height="499" alt="Image" src="https://github.com/user-attachments/assets/b4cd9ff7-0206-44bf-9768-da3b456a83fb" />
### Additional Information
_No response_
GiteaMirror
added the bug label 2026-05-13 07:48:53 -05:00
@AlbertoSinigaglia commented on GitHub (May 11, 2026):
I can confirm that that was the bug. I've added a prefix to the model name for VLLM removing the clash, and it worked... though it's looks like something that migh tbe quick to fix on the fronted code side, idk
<!-- gh-comment-id:4421726141 -->
@AlbertoSinigaglia commented on GitHub (May 11, 2026):
I can confirm that that was the bug. I've added a prefix to the model name for VLLM removing the clash, and it worked... though it's looks like something that migh tbe quick to fix on the fronted code side, idk
leaving open for triage, i agree it shouldn't crash due to model id clash.
<!-- gh-comment-id:4421809623 -->
@Classic298 commented on GitHub (May 11, 2026):
leaving open for triage, i agree it shouldn't crash due to model id clash.
Took a pass on this and opened a minimal fix in #24655.
Root cause appears to be in the chat model selector's virtualized dropdown: the rendered rows are keyed by the selector item value, which is not stable enough when multiple providers expose models that collapse to the same display/value alias. In that case Svelte throws each_key_duplicate and the dropdown fails to open.
The fix in #24655 keeps the behavior unchanged but switches the keyed identity for dropdown rows to prefer the backing model.id, with fallback only if needed. That should make the selector resilient when Ollama/vLLM (or other providers) surface overlapping model names.
If maintainers agree with the diagnosis, this should be a fairly contained frontend fix.
<!-- gh-comment-id:4440338494 -->
@linliang-bot commented on GitHub (May 13, 2026):
Took a pass on this and opened a minimal fix in #24655.
Root cause appears to be in the chat model selector's virtualized dropdown: the rendered rows are keyed by the selector item value, which is not stable enough when multiple providers expose models that collapse to the same display/value alias. In that case Svelte throws `each_key_duplicate` and the dropdown fails to open.
The fix in #24655 keeps the behavior unchanged but switches the keyed identity for dropdown rows to prefer the backing `model.id`, with fallback only if needed. That should make the selector resilient when Ollama/vLLM (or other providers) surface overlapping model names.
If maintainers agree with the diagnosis, this should be a fairly contained frontend fix.
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 @AlbertoSinigaglia on GitHub (May 11, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/24578
Check Existing Issues
Installation Method
Docker
Open WebUI Version
0.9.5
Ollama Version (if applicable)
Operating System
Browser (if applicable)
Chrome, but also tested on Safari
Confirmation
README.md.Expected Behavior
Actual Behavior
I always had a single ollama client connected to my OWUI instance, but since i've also added vLLM, I get:
Every time I try to open the dropdown menu for model selection
Steps to Reproduce
I think the problem is if the model name of vLLM is also present in Ollama, then they clash and break the whole model selectio menu.
So I guess that having an Ollama connection with model X, and a vLLM that exposes a model with the same X name will crash it correctly
Logs & Screenshots
Additional Information
No response
@AlbertoSinigaglia commented on GitHub (May 11, 2026):
I can confirm that that was the bug. I've added a prefix to the model name for VLLM removing the clash, and it worked... though it's looks like something that migh tbe quick to fix on the fronted code side, idk
@Classic298 commented on GitHub (May 11, 2026):
leaving open for triage, i agree it shouldn't crash due to model id clash.
@linliang-bot commented on GitHub (May 13, 2026):
Took a pass on this and opened a minimal fix in #24655.
Root cause appears to be in the chat model selector's virtualized dropdown: the rendered rows are keyed by the selector item value, which is not stable enough when multiple providers expose models that collapse to the same display/value alias. In that case Svelte throws
each_key_duplicateand the dropdown fails to open.The fix in #24655 keeps the behavior unchanged but switches the keyed identity for dropdown rows to prefer the backing
model.id, with fallback only if needed. That should make the selector resilient when Ollama/vLLM (or other providers) surface overlapping model names.If maintainers agree with the diagnosis, this should be a fairly contained frontend fix.