Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-21 01:25:28 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #5422] [BUG]: The desktop app on Windows is broken #5083

New Issue
Closed
opened 2026-06-05 14:51:54 -04:00 by yindo · 2 comments
yindo commented 2026-06-05 14:51:54 -04:00
Owner
Copy Link
Copy Source

Originally created by @michalpelszyk on GitHub (Apr 13, 2026).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5422

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Windows 24H2.
Installed the desktop app.

Unable to make it work.

Embedded models do not download thanks to Error downloading model: Internal server error.

Tried to import 2 GGUF files: failed with Error loading custom model. Code: 2.

Switching settings screen often ends up with infinite spinning circle. That is just traversing the settings sections, nothing else.

Unable to change to a different model (only tried embedded and LM Studio, really) in the UI. Not even with trying the OpenAI-compatible and providing the local LM Studio's address - just doesn't take it. It's only possible on fresh setup (deleted AppData/Roaming dir) and by direct editing the '.env' file.

When I did manage to set it up with LM Studio (see report 5421, the Ctrl+/ floating window only works on the FIRST invocation after ALLM start. After that the text input field is just impossible to activate: clicks, Tab key, arrows - nothing, it's dead. The screenshot ("+"), microphone and "@" buttons do work.

I am happy to provide any detail asked, do any troubleshooting (e.g. Process Monitor dumps), just tell me what to look at.

backend-2026-04-13.log

.env

Settings -> Event Log is empty, the few times I was able to get in there at all.

And here I wanted to start experimenting with local AIs...

Are there known steps to reproduce?

Install Anything LLM on Windows 11 (freshly installed 24H2).

Try to set it up initially: unable to download the model.

Click through despite that: Interface hangs.

Configure LM Studio (need to delete the folder from AppData/Roaming for that, unable to change config in the interface): Ctrl+/ only works once.

LLM Provider & Model (if applicable)

n/a

Embedder Provider & Model (if applicable)

Any

Originally created by @michalpelszyk on GitHub (Apr 13, 2026). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5422 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? Windows 24H2. Installed the desktop app. Unable to make it work. Embedded models do not download thanks to `Error downloading model: Internal server error.` Tried to import 2 GGUF files: failed with `Error loading custom model. Code: 2.` Switching settings screen often ends up with infinite spinning circle. That is just traversing the settings sections, nothing else. Unable to change to a different model (only tried embedded and LM Studio, really) in the UI. Not even with trying the OpenAI-compatible and providing the local LM Studio's address - just doesn't take it. It's only possible on fresh setup (deleted AppData/Roaming dir) and by direct editing the '.env' file. When I did manage to set it up with LM Studio (see report [5421](https://github.com/Mintplex-Labs/anything-llm/issues/5421), the Ctrl+/ floating window only works on the FIRST invocation after ALLM start. After that the text input field is just impossible to activate: clicks, Tab key, arrows - nothing, it's dead. The screenshot ("+"), microphone and "@" buttons do work. I am happy to provide any detail asked, do any troubleshooting (e.g. Process Monitor dumps), just tell me what to look at. [backend-2026-04-13.log](https://github.com/user-attachments/files/26674278/backend-2026-04-13.log.txt) [.env](https://github.com/user-attachments/files/26674287/env.txt) Settings -> Event Log is empty, the few times I was able to get in there at all. And here I wanted to start experimenting with local AIs... ### Are there known steps to reproduce? Install Anything LLM on Windows 11 (freshly installed 24H2). Try to set it up initially: unable to download the model. Click through despite that: Interface hangs. Configure LM Studio (need to delete the folder from AppData/Roaming for that, unable to change config in the interface): Ctrl+/ only works once. ### LLM Provider & Model (if applicable) n/a ### Embedder Provider & Model (if applicable) Any
yindo added the possible bugneeds info / can't replicate labels 2026-06-05 14:51:54 -04:00
yindo closed this issue 2026-06-05 14:51:54 -04:00
yindo commented 2026-06-05 14:51:55 -04:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Apr 13, 2026):

The logs show [AnythingLLMOllama] Boot failure for port 11434 repeatedly, meaning the embedded Ollama server isn't starting. This is almost certainly a local environment issue rather than a bug in AnythingLLM.

Please check:

  • Is anything else using port 11434? Run netstat -ano | findstr 11434 in PowerShell
  • Check Windows Defender/antivirus logs - the Ollama binary may be blocked or quarantined (it is signed so shouldnt be an issue)
  • Check Event Viewer (Windows Logs → Application) for any related errors
  • Try running AnythingLLM as Administrator

The "infinite spinning" and UI issues are downstream effects of the LLM backend failing to connect. That being said, something is up with your client device as you are the only person reporting this and Windows is majority of our userbase. If the app was this level of broken I would have 10K emails in my inbox.

Can you tell me more about your environment (using Proxmox, on a VM, etc). What kind of machine are you on? I see youre just trying to run llama 3.2 3B - which is quite small so shouldnt be a device spec issue.

<!-- gh-comment-id:4238171935 --> @timothycarambat commented on GitHub (Apr 13, 2026): The logs show `[AnythingLLMOllama] Boot failure for port 11434` repeatedly, meaning the embedded Ollama server isn't starting. This is almost certainly a local environment issue rather than a bug in AnythingLLM. Please check: - Is anything else using port 11434? Run netstat -ano | findstr 11434 in PowerShell - Check Windows Defender/antivirus logs - the Ollama binary may be blocked or quarantined (it is signed so shouldnt be an issue) - Check Event Viewer (Windows Logs → Application) for any related errors - Try running AnythingLLM as Administrator The "infinite spinning" and UI issues are downstream effects of the LLM backend failing to connect. That being said, something is up with your client device as you are the only person reporting this and Windows is majority of our userbase. If the app was this level of broken I would have 10K emails in my inbox. Can you tell me more about your environment (using Proxmox, on a VM, etc). What kind of machine are you on? I see youre just trying to run llama 3.2 3B - which is quite small so shouldnt be a device spec issue.
yindo commented 2026-06-05 14:51:55 -04:00
Author
Owner
Copy Link
Copy Source

@michalpelszyk commented on GitHub (Apr 14, 2026):

Dear @timothycarambat,

Thank you for the reply.

  • netstat -ano | findstr 11434 shows nothing.
  • Defender logs don't have anything about either AnythingLLM or Ollama. Only info about completed scans.
  • Nothing in Windows Event Logs either.
  • Running as administrator didn't change anything - same.

Machine: Windows 11 24h2, installed last month because Docker Desktop refused to install on 10.
Bare metal.
Ryzen 5600x, 32GB RAM, RTX 5060.
Installed Docker Desktop, WSL, LM Studio, Steam, Epic, Sunshine and UltraVNC.

Backend seems to be able to run for a short while - I was able to edit and save a default system prompt shortly after ALLM start. After 3-5 minutes of it running, though, it's back to the Failed to fetch error.

<!-- gh-comment-id:4247579437 --> @michalpelszyk commented on GitHub (Apr 14, 2026): Dear @timothycarambat, Thank you for the reply. * `netstat -ano | findstr 11434 ` shows nothing. * Defender logs don't have anything about either AnythingLLM or Ollama. Only info about completed scans. * Nothing in Windows Event Logs either. * Running as administrator didn't change anything - same. Machine: Windows 11 24h2, installed last month because Docker Desktop refused to install on 10. Bare metal. Ryzen 5600x, 32GB RAM, RTX 5060. Installed Docker Desktop, WSL, LM Studio, Steam, Epic, Sunshine and UltraVNC. Backend seems to be able to run for a short while - I was able to edit and save a default system prompt shortly after ALLM start. After 3-5 minutes of it running, though, it's back to the `Failed to fetch` error.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label needs info / can't replicate possible bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#5083
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?