WARN Constants Failed to get LMStudio models: fetch failed

[Arav-Shakya]

Describe the bug

I tried to run bolt.diy on my VM and it's giving me this WARN Constants Failed to get LMStudio models: fetch failed every time I give a prompt. I am use Ollama locally

Link to the Bolt URL that caused the error

http://localhost:5173/chat/4

Steps to reproduce

1.Create a VM on google cloud. ( Use L4 GPU, Machine type : g2-standard-16, 64gb memory )
2.Select windows server 2019
3.Install Latest Node.js
4.Git bash and VS code
5.Install Ollama Qwen 2.5 Coder 32B
6.Install node module and run repo.
7. change OLLAMA_API_BASE_URL=http://127.0.0.1:11434 in .env

Expected behavior

once you will try to select the Ollama model and give it a prompt it will give WARN Constants Failed to get LMStudio models: fetch failed.

Screen Recording / Screenshot

No response

Platform

Browser: Chrome
OS: Windows Server 2019 with desktop environment.
VM: Google Cloud

Provider Used

No response

Model Used

No response

Additional context

No response

[Mhinolv]

Thank you for submitting this, I am seeing the same thing.

[domforson]

I’m experiencing the same thing even though I use Gemini models. I’ve never used Ollama but I’m getting the same alert.

[dustinwloring1988]

@domforson @Mhinolv @Arav-Shakya I tried the latest version and turned LM Studio off in the settings menu and this fixed it.

[libbi3605]

yeah no chance to take full advantage of it if it dont support it. i was wondering if i run ollama in docker too might work. also not sure but i think it worked before and dont work now.. thanks anyone for advice...

[lamanweb]

change your browser, or clear cookies / data.

[Arav-Shakya]

Thank you for submitting this, I am seeing the same thing.

I did find a solution to this: Go to the .env.example file change it to .env.local if you are using Ollama then change the following:

Before:
OLLAMA_API_BASE_URL=
After
OLLAMA_API_BASE_URL=http://127.0.0.1:11434

This one below will solve LMS issue
Before:
LMSTUDIO_API_BASE_URL =
After:
LMSTUDIO_API_BASE_URL=http:/127.0.0.1:1234

After these changes issues were resolved but I tried running Qwen 2.5 coder 32b model but I didn't got any response or any error.

pls see if it works for you or not.

[MishaNyaCopilot]

Changing .env doesnt work for me

[VictimOfPing]

Hi I have a problem when I go to use Ollama, when I go to send it a message it gives me this error: There was an error processing your request: An error occurred.

[roodfps]

i'm having the same issue, some one know how to solve it?

[Arav-Shakya]

Provide more context with screenshot.

[thecodacus]

for ollama please use the UI to set the base url
and try again

from here

[VictimOfPing]

Provide more context with screenshot.

[thecodacus]

can you show the terminal screenshot ..
where you have put the pnpm run dev

[VictimOfPing]

can you show the terminal screenshot .. where you have put the pnpm run dev

is empty

[thecodacus]

can you try if this PR works for you
#816