[GH-ISSUE #24544] issue: llamacpp load/unload indicator doesn't work #91065

New Issue

GiteaMirror · 2026-05-15T16:20:07-05:00

GiteaMirror commented

2026-05-15 16:20:07 -05:00

Originally created by @frenzybiscuit on GitHub (May 10, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/24544

Check Existing Issues

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

Git Clone

Open WebUI Version

0.9.4

Ollama Version (if applicable)

N/A

Operating System

Linux

Browser (if applicable)

Firefox

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

https://docs.openwebui.com/getting-started/quick-start/connect-a-provider/starting-with-llama-cpp/#unloading-loaded-models

Per this document loaded models should have a green icon next to them indicating they are loaded.

This does not happen on router mode with llamacpp.

And I have no idea how this feature would work without router mode. However, even when launching with regular llama-server and no router mode, the icon doesn't show up.

Actual Behavior

Doesn't work

Steps to Reproduce

See screenshots along with the below.

config.ini:

[Default]
model = gguf/Gemsicle-31B-Q8_0-hb16.gguf
n-gpu-layers = 999
ctx-size = 400000
split-mode = tensor
ctx-checkpoints = 4
cache-ram = 80960
parallel = 4
flash-attn = on
reasoning = on
reasoning-budget = 1024
reasoning-budget-message = "... thinking budget exceeded, let's answer now."

[Gemsicle | 31B | Q8 | HB16]
model = gguf/Gemsicle-31B-Q8_0-hb16.gguf
n-gpu-layers = 999
ctx-size = 400000
split-mode = tensor
ctx-checkpoints = 4
cache-ram = 80960
parallel = 4
flash-attn = on
reasoning = on
reasoning-budget = 1024
reasoning-budget-message = "... thinking budget exceeded, let's answer now."

llama.cpp-b9097/build/bin/llama-server --models-max 1 --sleep-idle-seconds 1200 --api-key keyhere --host 127.0.0.1 --port 5000 --models-preset ./config.ini --slots --metrics

Logs & Screenshots

Additional Information

N/A

Originally created by @frenzybiscuit on GitHub (May 10, 2026). Original GitHub issue: https://github.com/open-webui/open-webui/issues/24544 ### 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 Git Clone ### Open WebUI Version 0.9.4 ### Ollama Version (if applicable) N/A ### Operating System Linux ### Browser (if applicable) Firefox ### 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 https://docs.openwebui.com/getting-started/quick-start/connect-a-provider/starting-with-llama-cpp/#unloading-loaded-models Per this document loaded models should have a green icon next to them indicating they are loaded. This does not happen on router mode with llamacpp. And I have no idea how this feature would work without router mode. However, even when launching with regular llama-server and no router mode, the icon doesn't show up. ### Actual Behavior Doesn't work ### Steps to Reproduce See screenshots along with the below. config.ini: ``` [Default] model = gguf/Gemsicle-31B-Q8_0-hb16.gguf n-gpu-layers = 999 ctx-size = 400000 split-mode = tensor ctx-checkpoints = 4 cache-ram = 80960 parallel = 4 flash-attn = on reasoning = on reasoning-budget = 1024 reasoning-budget-message = "... thinking budget exceeded, let's answer now." [Gemsicle | 31B | Q8 | HB16] model = gguf/Gemsicle-31B-Q8_0-hb16.gguf n-gpu-layers = 999 ctx-size = 400000 split-mode = tensor ctx-checkpoints = 4 cache-ram = 80960 parallel = 4 flash-attn = on reasoning = on reasoning-budget = 1024 reasoning-budget-message = "... thinking budget exceeded, let's answer now." ``` ```llama.cpp-b9097/build/bin/llama-server --models-max 1 --sleep-idle-seconds 1200 --api-key keyhere --host 127.0.0.1 --port 5000 --models-preset ./config.ini --slots --metrics``` ### Logs & Screenshots <img width="1040" height="1056" alt="Image" src="https://github.com/user-attachments/assets/1184aac2-a98d-47e9-aa73-ade68a11d645" /> ### Additional Information N/A

GiteaMirror added the bug label 2026-05-15 16:20:07 -05:00

GiteaMirror closed this issue

2026-05-15 16:20:08 -05:00

GiteaMirror commented

2026-05-15 16:20:08 -05:00

@frenzybiscuit commented on GitHub (May 10, 2026):

I see this is fixed in dev now. I'll try it when the next update is released as I don't use the dev branch.

Thanks!

@frenzybiscuit commented on GitHub (May 10, 2026): I see this is fixed in dev now. I'll try it when the next update is released as I don't use the dev branch. Thanks!

Sign in to join this conversation.

Branches Tags

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/open-webui#91065