[GH-ISSUE #3216] [BUG]: File integration with LM Studio not working #2070

Closed
opened 2026-02-22 18:28:01 -05:00 by yindo · 1 comment
Owner

Originally created by @Raphango on GitHub (Feb 14, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3216

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Image

When I attach a file, the interface behaves like a file was correctly sent to LM Studio. However, my LLM behaves as if no file was provided. Maybe the integration with LM Studio should enable further configurations for file processing and RAG alike.

Are there known steps to reproduce?

  1. Start LM Studio Local Server
  2. Start AnythingLLM
  3. Connect to LM Studio Local Server
  4. Test if the connection/responsa is alive
  5. Upload file and pin it
  6. Ask the LLM about the file.
Originally created by @Raphango on GitHub (Feb 14, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3216 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? ![Image](https://github.com/user-attachments/assets/12b45e12-6ebe-45e8-b550-aa5c06abf3de) When I attach a file, the interface behaves like a file was correctly sent to LM Studio. However, my LLM behaves as if no file was provided. Maybe the integration with LM Studio should enable further configurations for file processing and RAG alike. ### Are there known steps to reproduce? 1. Start LM Studio Local Server 2. Start AnythingLLM 3. Connect to LM Studio Local Server 4. Test if the connection/responsa is alive 5. Upload file and pin it 6. Ask the LLM about the file.
yindo added the possible bug label 2026-02-22 18:28:01 -05:00
yindo closed this issue 2026-02-22 18:28:01 -05:00
Author
Owner

@timothycarambat commented on GitHub (Feb 14, 2025):

We for sure send the content since it is a citation in the response. There is no two-ways about that. Are you using a deepseek distillation? They are quite bad at the <32B param of understanding content injection.

Also:

Upload file and pin it

The document may still be being pruned from the response. We still adhere to the model context window since if we let it overflow the LLM will crash.

@timothycarambat commented on GitHub (Feb 14, 2025): We for sure send the content since it is a citation in the response. There is no two-ways about that. Are you using a deepseek distillation? They are quite bad at the <32B param of understanding content injection. Also: > Upload file and pin it The document may still be being pruned from the response. We still adhere to the model context window since if we let it overflow the LLM will crash.
yindo changed title from [BUG]: File integration with LM Studio not working to [GH-ISSUE #3216] [BUG]: File integration with LM Studio not working 2026-06-05 14:44:20 -04:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#2070