I have read and followed all the instructions provided in the README.md.
I have reviewed the troubleshooting.md document.
I have included the browser console logs.
I have included the Docker container logs.
Logs and Screenshots
Browser Console Logs:
Array []
initNewChat
<empty string>. <<< here should be the modelfile logged out I think
Object { status: true, version: "v1.0.0-alpha.92", auth: true, default_models: null, default_prompt_suggestions: (4) […] }
Docker Container Logs:
N/A
Screenshots (if applicable):
N/A
Installation Method
Installed with docker based on readme
Additional Information
N/A
Note
If the bug report is incomplete or does not follow the provided instructions, it may not be addressed. Please ensure that you have followed the steps outlined in the README.md and troubleshooting.md documents, and provide all necessary information for us to reproduce and address the issue. Thank you!
Originally created by @keriati on GitHub (Feb 6, 2024).
# Bug Report
It seems the modelfiles are not loaded on the chat page.
## Description
It seems the modelfiles are not loaded on the chat page.
**Bug Summary:**
It seems the modelfiles are not loaded on the chat page.
**Steps to Reproduce:**
- Go to ollama hub
- Add based dolphin mixtral
- Adjust the content so it starts with:
```
FROM dolphin-mixtral:8x7b-v2.7-q5_K_M
```
- Save it
- on the modelfiles page click on Based Dolphin Mixtral
- get reidrected to chat page
**Expected Behavior:**
- chat page applies the modelfile settings
**Actual Behavior:**
- chat page does not apply the modelfile settings
## Environment
- **Operating System:** [e.g., Windows 10, macOS Big Sur, Ubuntu 20.04]
macos
- **Browser (if applicable):** [e.g., Chrome 100.0, Firefox 98.0]
latest chrome
latest firefox
## Reproduction Details
**Confirmation:**
- [x] I have read and followed all the instructions provided in the README.md.
- [x] I have reviewed the troubleshooting.md document.
- [x] I have included the browser console logs.
- [x] I have included the Docker container logs.
## Logs and Screenshots
**Browser Console Logs:**
```
Array []
initNewChat
<empty string>. <<< here should be the modelfile logged out I think
Object { status: true, version: "v1.0.0-alpha.92", auth: true, default_models: null, default_prompt_suggestions: (4) […] }
```
**Docker Container Logs:**
N/A
**Screenshots (if applicable):**
N/A
## Installation Method
Installed with docker based on readme
## Additional Information
N/A
## Note
If the bug report is incomplete or does not follow the provided instructions, it may not be addressed. Please ensure that you have followed the steps outlined in the README.md and troubleshooting.md documents, and provide all necessary information for us to reproduce and address the issue. Thank you!
Unable to reproduce the issue, could you verify that you're on the latest version of the webui?
@tjbck commented on GitHub (Feb 6, 2024):
Unable to reproduce the issue, could you verify that you're on the latest version of the webui?

Can it be the problem that the FROM part is adjusted?
@keriati commented on GitHub (Feb 6, 2024):
Sure:

Can it be the problem that the FROM part is adjusted?
You're saying can't get to the second part of the screenshot correct?
Does the model at least show up in the model selector?
@tjbck commented on GitHub (Feb 6, 2024):


You're saying can't get to the second part of the screenshot correct?

Does the model at least show up in the model selector?
@keriati commented on GitHub (Feb 6, 2024):
Model selection:

Can you manually select the model? As for the modelfiles page, the url should look something like this for the model: http://localhost:3000/?models=[your model name]
@tjbck commented on GitHub (Feb 6, 2024):
Can you manually select the model? As for the modelfiles page, the url should look something like this for the model: `http://localhost:3000/?models=[your model name]`
thanks for the hint, I think I see now what is going on:
This is added to the URL when I click on the modefile page and it does not work:
models=based-dolphin-mixtral%3Alatest
this however works:
?models=dolphin-mixtral:8x7b-v2.7-q5_K_M
(works = the model is selected, however the system prompt is not applied.)
@keriati commented on GitHub (Feb 6, 2024):
thanks for the hint, I think I see now what is going on:
This is added to the URL when I click on the modefile page and it does not work:
```models=based-dolphin-mixtral%3Alatest```
this however works:
```?models=dolphin-mixtral:8x7b-v2.7-q5_K_M```
(works = the model is selected, however the system prompt is not applied.)
Did you also change the model tag name somehow? Could you provide us with the step by step guide with images on how we can reproduce the issue?
@tjbck commented on GitHub (Feb 6, 2024):
Did you also change the model tag name somehow? Could you provide us with the step by step guide with images on how we can reproduce the issue?
So I only adjusted the FROM part of the model as I needed the bigger version:
@keriati commented on GitHub (Feb 6, 2024):
So I only adjusted the FROM part of the model as I needed the bigger version:

okay, so I pressed save and update now again and now it works
@keriati commented on GitHub (Feb 6, 2024):
okay, so I pressed save and update now again and now it works

Also now in the model selection i get this based-dolphin-mixtral:latest instead of dolphin-mixtral:8x7b-v2.7-q5_K_M
@keriati commented on GitHub (Feb 6, 2024):
Also now in the model selection i get this based-dolphin-mixtral:latest instead of dolphin-mixtral:8x7b-v2.7-q5_K_M
Strange, that should've been the case in the first place. Maybe there was a connection issue with your Ollama instance. I'll close this for now but feel free to ping me if you encounter the same issue again, Thanks!
@tjbck commented on GitHub (Feb 6, 2024):
Strange, that should've been the case in the first place. Maybe there was a connection issue with your Ollama instance. I'll close this for now but feel free to ping me if you encounter the same issue again, Thanks!
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 @keriati on GitHub (Feb 6, 2024).
Bug Report
It seems the modelfiles are not loaded on the chat page.
Description
It seems the modelfiles are not loaded on the chat page.
Bug Summary:
It seems the modelfiles are not loaded on the chat page.
Steps to Reproduce:
Expected Behavior:
Actual Behavior:
Environment
Operating System: [e.g., Windows 10, macOS Big Sur, Ubuntu 20.04]
macos
Browser (if applicable): [e.g., Chrome 100.0, Firefox 98.0]
latest chrome
latest firefox
Reproduction Details
Confirmation:
Logs and Screenshots
Browser Console Logs:
Docker Container Logs:
N/A
Screenshots (if applicable):
N/A
Installation Method
Installed with docker based on readme
Additional Information
N/A
Note
If the bug report is incomplete or does not follow the provided instructions, it may not be addressed. Please ensure that you have followed the steps outlined in the README.md and troubleshooting.md documents, and provide all necessary information for us to reproduce and address the issue. Thank you!
@tjbck commented on GitHub (Feb 6, 2024):
Unable to reproduce the issue, could you verify that you're on the latest version of the webui?
@keriati commented on GitHub (Feb 6, 2024):
Sure:
Can it be the problem that the FROM part is adjusted?
@tjbck commented on GitHub (Feb 6, 2024):
You're saying can't get to the second part of the screenshot correct?
Does the model at least show up in the model selector?
@keriati commented on GitHub (Feb 6, 2024):
Modelfiles page:

Click on Based Dolphin result:

@keriati commented on GitHub (Feb 6, 2024):
Model selection:
@tjbck commented on GitHub (Feb 6, 2024):
Can you manually select the model? As for the modelfiles page, the url should look something like this for the model:
http://localhost:3000/?models=[your model name]@keriati commented on GitHub (Feb 6, 2024):
thanks for the hint, I think I see now what is going on:
This is added to the URL when I click on the modefile page and it does not work:
models=based-dolphin-mixtral%3Alatestthis however works:
?models=dolphin-mixtral:8x7b-v2.7-q5_K_M(works = the model is selected, however the system prompt is not applied.)
@tjbck commented on GitHub (Feb 6, 2024):
Did you also change the model tag name somehow? Could you provide us with the step by step guide with images on how we can reproduce the issue?
@keriati commented on GitHub (Feb 6, 2024):
So I only adjusted the FROM part of the model as I needed the bigger version:
@tjbck commented on GitHub (Feb 6, 2024):
And if you save and update, does that just take you back to the modelfiles page?
@keriati commented on GitHub (Feb 6, 2024):
okay, so I pressed save and update now again and now it works
@keriati commented on GitHub (Feb 6, 2024):
Also now in the model selection i get this based-dolphin-mixtral:latest instead of dolphin-mixtral:8x7b-v2.7-q5_K_M
@tjbck commented on GitHub (Feb 6, 2024):
Strange, that should've been the case in the first place. Maybe there was a connection issue with your Ollama instance. I'll close this for now but feel free to ping me if you encounter the same issue again, Thanks!