[GH-ISSUE #1726] [BUG]: Unexpected Model Switch Issue #1126

Closed
opened 2026-02-22 18:23:13 -05:00 by yindo · 3 comments
Owner

Originally created by @chalitbkb on GitHub (Jun 19, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/1726

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

I'm confident that I'm using the Gemini1.5 Pro model and that I'm in the same workspace as this model, which can handle large prompts. So why is it like in the picture? It's strange that the model is automatically selected as OpenAI.

image
image

Are there known steps to reproduce?

No response

Originally created by @chalitbkb on GitHub (Jun 19, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/1726 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? I'm confident that I'm using the Gemini1.5 Pro model and that I'm in the same workspace as this model, which can handle large prompts. So why is it like in the picture? It's strange that the model is automatically selected as OpenAI. ![image](https://github.com/Mintplex-Labs/anything-llm/assets/32437546/953bff85-1b81-4f44-aaf2-f150c4b562dd) ![image](https://github.com/Mintplex-Labs/anything-llm/assets/32437546/86bfb5aa-be12-4dbf-9447-c4c035f24853) ### Are there known steps to reproduce? _No response_
yindo added the possible bug label 2026-02-22 18:23:13 -05:00
yindo closed this issue 2026-02-22 18:23:14 -05:00
Author
Owner

@timothycarambat commented on GitHub (Jun 19, 2024):

OpenAI failed to embed

Is OpenAI you embedder? In order to do vector search the prompt is embedded and then searched in the vector DB. it appears your prompt is too long for that embedders context window.

This is also known as well

@timothycarambat commented on GitHub (Jun 19, 2024): > OpenAI failed to embed Is OpenAI you embedder? In order to do vector search the prompt is embedded and then searched in the vector DB. it appears your prompt is too long for that embedders context window. This is also known as well
Author
Owner

@chalitbkb commented on GitHub (Jun 19, 2024):

OpenAI failed to embed

Is OpenAI you embedder? In order to do vector search the prompt is embedded and then searched in the vector DB. it appears your prompt is too long for that embedders context window.

This is also known as well

How can we solve this problem, or are there any alternatives that can handle a large prompt? I want to use Gemini.

@timothycarambat

@chalitbkb commented on GitHub (Jun 19, 2024): > > OpenAI failed to embed > > Is OpenAI you embedder? In order to do vector search the prompt is embedded and then searched in the vector DB. it appears your prompt is too long for that embedders context window. > > This is also known as well How can we solve this problem, or are there any alternatives that can handle a large prompt? I want to use Gemini. @timothycarambat
Author
Owner

@timothycarambat commented on GitHub (Jun 19, 2024):

@chalitbkb Currently, you can actually just chat with a workspace with no documents. A very large prompt would likely introduce very bad vector search results

@timothycarambat commented on GitHub (Jun 19, 2024): @chalitbkb Currently, you can actually just chat with a workspace with no documents. A very large prompt would likely introduce very bad vector search results
yindo changed title from [BUG]: Unexpected Model Switch Issue to [GH-ISSUE #1726] [BUG]: Unexpected Model Switch Issue 2026-06-05 14:39:07 -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#1126