Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 22:23:50 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #2259] [BUG]: Ollama call failed with status code 500: llama runner process has terminated: signal: segmentation fault It is on a Mac Sillicon #1475

New Issue
Closed
opened 2026-02-22 18:24:59 -05:00 by yindo · 6 comments
yindo commented 2026-02-22 18:24:59 -05:00
Owner
Copy Link
Copy Source

Originally created by @venturaEffect on GitHub (Sep 10, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2259

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Even if there is no Ollama selected and it is an imported LLM I get this message when trying to chat:

Ollama call failed with status code 500: llama runner process has terminated: signal: segmentation fault
It is on a Mac Sillicon

Appreciate your time.

Are there known steps to reproduce?

Just installed fresh on a Mac (Apple Sillicon) the AnythingLLM desktop. Selected the imported LLM with the AnythingLLM provider, the Embedder is also AnythingLLM and LanceDB on the VectorDB.

Also on the settings from the workspace is Workspace Agent LLM Provider: AnythingLLM and the Workspace Agent model is the custom model: llama-3-Instruct-8b-...

Originally created by @venturaEffect on GitHub (Sep 10, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2259 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? Even if there is no Ollama selected and it is an imported LLM I get this message when trying to chat: Ollama call failed with status code 500: llama runner process has terminated: signal: segmentation fault It is on a Mac Sillicon Appreciate your time. ### Are there known steps to reproduce? Just installed fresh on a Mac (Apple Sillicon) the AnythingLLM desktop. Selected the imported LLM with the AnythingLLM provider, the Embedder is also AnythingLLM and LanceDB on the VectorDB. Also on the settings from the workspace is Workspace Agent LLM Provider: AnythingLLM and the Workspace Agent model is the custom model: llama-3-Instruct-8b-...
yindo added the possible bug label 2026-02-22 18:24:59 -05:00
yindo closed this issue 2026-02-22 18:25:00 -05:00
yindo commented 2026-02-22 18:25:00 -05:00
Author
Owner
Copy Link
Copy Source

@shatfield4 commented on GitHub (Sep 10, 2024):

When you use the AnythingLLM provider, this actually uses Ollama under the hood to run the models which is why you are getting Ollama errors.

After you set AnythingLLM as the LLM provider, select a model, and press save do you see the model start to download in the top right corner of the app?

@shatfield4 commented on GitHub (Sep 10, 2024): When you use the AnythingLLM provider, this actually uses Ollama under the hood to run the models which is why you are getting Ollama errors. After you set AnythingLLM as the LLM provider, select a model, and press save do you see the model start to download in the top right corner of the app?
yindo commented 2026-02-22 18:25:00 -05:00
Author
Owner
Copy Link
Copy Source

@venturaEffect commented on GitHub (Sep 10, 2024):

Hi, we are not using Ollama. We are importing one custom model. And after it is installed we click on save. We did that several times and closed and open the app again. But still get this error.

@venturaEffect commented on GitHub (Sep 10, 2024): Hi, we are not using Ollama. We are importing one custom model. And after it is installed we click on save. We did that several times and closed and open the app again. But still get this error.
yindo commented 2026-02-22 18:25:00 -05:00
Author
Owner
Copy Link
Copy Source

@venturaEffect commented on GitHub (Sep 10, 2024):

When loading a model from Ollama that are by default on AnythingLLM it seems to work but the problem is by trying the custom model. This is strange because the same model on my Windows11 with AnythingLLM desktop also installed works just fine. But on the Apple M1 not.

Do you know why this could be?

Appreciate.

@venturaEffect commented on GitHub (Sep 10, 2024): When loading a model from Ollama that are by default on AnythingLLM it seems to work but the problem is by trying the custom model. This is strange because the same model on my Windows11 with AnythingLLM desktop also installed works just fine. But on the Apple M1 not. Do you know why this could be? Appreciate.
yindo commented 2026-02-22 18:25:00 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Sep 10, 2024):

to clarify: What @shatfield4 is saying is that the default LLM provider inside of AnythingLLM is Ollama. So when you import the model into AnythingLLM and run it, it is using the pre-packaged Ollama inside of AnythingLLM.

The model being used is simply GGUF correct? For the model to work on Windows but not on Mac would indicate to me initially that something went wrong with the file transfer/download ending up with a corrupted or incomplete file. Since GGUF is a portable format, it should work fine across OS as the ollama version in all desktop instances are consistent.

Closing as unable to reproduce for now, but keep the thread open to continue discussion. Are there any debug logs you can pull at the time of crash?

@timothycarambat commented on GitHub (Sep 10, 2024): to clarify: What @shatfield4 is saying is that the _default LLM provider_ inside of AnythingLLM is Ollama. So when you import the model into AnythingLLM and run it, it is using the pre-packaged Ollama inside of AnythingLLM. The model being used is simply GGUF correct? For the model to work on Windows but not on Mac would indicate to me initially that something went wrong with the file transfer/download ending up with a corrupted or incomplete file. Since GGUF is a portable format, it should work fine across OS as the ollama version in all desktop instances are consistent. Closing as unable to reproduce for now, but keep the thread open to continue discussion. Are there any [debug logs](https://docs.anythingllm.com/debug) you can pull at the time of crash?
yindo commented 2026-02-22 18:25:00 -05:00
Author
Owner
Copy Link
Copy Source

@venturaEffect commented on GitHub (Sep 10, 2024):

Ok, I'm going to download the .gguf file from HuggingFace and import it again to AnythingLLM.

Thanks for the advice.

@venturaEffect commented on GitHub (Sep 10, 2024): Ok, I'm going to download the .gguf file from HuggingFace and import it again to AnythingLLM. Thanks for the advice.
yindo commented 2026-02-22 18:25:00 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Sep 10, 2024):

Can you link the repo to the model you wish to use? Will help reproduce if it is public

@timothycarambat commented on GitHub (Sep 10, 2024): Can you link the repo to the model you wish to use? Will help reproduce if it is public
yindo changed title from [BUG]: Ollama call failed with status code 500: llama runner process has terminated: signal: segmentation fault It is on a Mac Sillicon to [GH-ISSUE #2259] [BUG]: Ollama call failed with status code 500: llama runner process has terminated: signal: segmentation fault It is on a Mac Sillicon 2026-06-05 14:40:58 -04:00
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 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#1475
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?