Docker Engine on Linux via "sudo ollama serve" followed by "sudo docker run -d -p 3000:8080 --gpus=all -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollama
Environment
Open WebUI Version: 0.4.8
Ollama (if applicable): 0.5.1
Operating System: Ubuntu 24.04.1 LTS
Browser (if applicable): Firefox
Expected Behavior:
Based on what I saw in the docs (see here: https://docs.openwebui.com/features/#-model-management ) I should be able to select huggingface models or local GGUF files from going to a "experimental" section in the models tab under the admin settings. Specifically, the documentation says this:
"⬆️ GGUF File Model Creation: Effortlessly create Ollama models by uploading GGUF files directly from Open WebUI from the Admin Settings > Settings > Model > Experimental menu. The process has been streamlined with the option to upload from your machine or download GGUF files from Hugging Face."
Actual Behavior:
There is no experimental menu, nor does simply doing something like putting the huggingface model url into the models searchbar to pull the model working. For the model searchbar option, I chose the example: huggingface.co/chaoticneutrals/tensorblock/hathor_respawn-l3-8b-v0.8-gguf:q4_k_m and trying to pull it doesn't work, see screenshot (gives the "download cancelled" error). I do not see any way to grab a local GGUF file.
Logs and Screenshots
Originally created by @rchaklashiya on GitHub (Dec 22, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/7994
## Installation Method
Docker Engine on Linux via "sudo ollama serve" followed by "`sudo docker run -d -p 3000:8080 --gpus=all -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollama`
## Environment
- **Open WebUI Version:** 0.4.8
- **Ollama (if applicable):** 0.5.1
- **Operating System:** Ubuntu 24.04.1 LTS
- **Browser (if applicable):** Firefox
## Expected Behavior:
Based on what I saw in the docs (see here: https://docs.openwebui.com/features/#-model-management ) I should be able to select huggingface models or local GGUF files from going to a "experimental" section in the models tab under the admin settings. Specifically, the documentation says this:
"⬆️ GGUF File Model Creation: Effortlessly create Ollama models by uploading GGUF files directly from Open WebUI from the Admin Settings > Settings > Model > Experimental menu. The process has been streamlined with the option to upload from your machine or download GGUF files from Hugging Face."
## Actual Behavior:
There is no experimental menu, nor does simply doing something like putting the huggingface model url into the models searchbar to pull the model working. For the model searchbar option, I chose the example: huggingface.co/chaoticneutrals/tensorblock/hathor_respawn-l3-8b-v0.8-gguf:q4_k_m and trying to pull it doesn't work, see screenshot (gives the "download cancelled" error). I do not see any way to grab a local GGUF file.
## Logs and Screenshots
@JohnathanCrow commented on GitHub (Dec 22, 2024):
Not sure about pulling direct, but for local: Admin Settings > Connections > Manage Ollama Connections (Spanner Icon) > Experimental (Show)
Some of the recent UI changes have been a little confusing for me too. I'd already found the new location for upload before, but it still took me a minute to remember. I think I read that the docs are kind of in progress, so some things are probably outdated.
<!-- gh-comment-id:2558317146 -->
@JohnathanCrow commented on GitHub (Dec 22, 2024):
Not sure about pulling direct, but for local: Admin Settings > Connections > Manage Ollama Connections (Spanner Icon) > Experimental (Show)
Some of the recent UI changes have been a little confusing for me too. I'd already found the new location for upload before, but it still took me a minute to remember. I think I read that the docs are kind of in progress, so some things are probably outdated.
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 @rchaklashiya on GitHub (Dec 22, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/7994
Installation Method
Docker Engine on Linux via "sudo ollama serve" followed by "
sudo docker run -d -p 3000:8080 --gpus=all -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollamaEnvironment
Open WebUI Version: 0.4.8
Ollama (if applicable): 0.5.1
Operating System: Ubuntu 24.04.1 LTS
Browser (if applicable): Firefox
Expected Behavior:
Based on what I saw in the docs (see here: https://docs.openwebui.com/features/#-model-management ) I should be able to select huggingface models or local GGUF files from going to a "experimental" section in the models tab under the admin settings. Specifically, the documentation says this:
"⬆️ GGUF File Model Creation: Effortlessly create Ollama models by uploading GGUF files directly from Open WebUI from the Admin Settings > Settings > Model > Experimental menu. The process has been streamlined with the option to upload from your machine or download GGUF files from Hugging Face."
Actual Behavior:
There is no experimental menu, nor does simply doing something like putting the huggingface model url into the models searchbar to pull the model working. For the model searchbar option, I chose the example: huggingface.co/chaoticneutrals/tensorblock/hathor_respawn-l3-8b-v0.8-gguf:q4_k_m and trying to pull it doesn't work, see screenshot (gives the "download cancelled" error). I do not see any way to grab a local GGUF file.
Logs and Screenshots
@JohnathanCrow commented on GitHub (Dec 22, 2024):
Not sure about pulling direct, but for local: Admin Settings > Connections > Manage Ollama Connections (Spanner Icon) > Experimental (Show)
Some of the recent UI changes have been a little confusing for me too. I'd already found the new location for upload before, but it still took me a minute to remember. I think I read that the docs are kind of in progress, so some things are probably outdated.