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 #3029] [FEAT]: easier method to choose provider / models (desktop) #1938

New Issue
Closed
opened 2026-02-22 18:27:19 -05:00 by yindo · 3 comments
yindo commented 2026-02-22 18:27:19 -05:00
Owner
Copy Link
Copy Source

Originally created by @ozp on GitHub (Jan 25, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3029

What would you like to see?

I found that it is confusing to choose provider / model at the desktop version.
Its ok to have the full list to choose and add your key.
But at the workspace, it should apear only the provider / models that you previously selected.

Originally created by @ozp on GitHub (Jan 25, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3029 ### What would you like to see? I found that it is confusing to choose provider / model at the desktop version. Its ok to have the full list to choose and add your key. But at the workspace, it should apear only the provider / models that you previously selected.
yindo added the enhancementfeature request labels 2026-02-22 18:27:19 -05:00
yindo closed this issue 2026-02-22 18:27:19 -05:00
yindo commented 2026-02-22 18:27:19 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jan 27, 2025):

Can you explain more? When the workspace provider is selected and it has not yet been defined we ask you to set up that provider right there. This is not in desktop yet, but was done via https://github.com/Mintplex-Labs/anything-llm/pull/2865

@timothycarambat commented on GitHub (Jan 27, 2025): Can you explain more? When the workspace provider is selected and it has not yet been defined we ask you to set up that provider right there. This is not in desktop yet, but was done via https://github.com/Mintplex-Labs/anything-llm/pull/2865
yindo commented 2026-02-22 18:27:19 -05:00
Author
Owner
Copy Link
Copy Source

@ozp commented on GitHub (Jan 29, 2025):

ok, its more a UX issue

I'm not good at english, so I asked deepseek for help:

Feature Request: Simplified Provider/Model Selection in Desktop Version

Description:
The current process for selecting providers and models in the desktop version can be confusing and cumbersome, especially for regular users. While the full list of providers and models is necessary for initial setup and key configuration, the workspace experience should be streamlined to only show the providers and models that have been previously selected and configured by the user.

Current Workflow:

  1. Settings: Users can choose providers, models, and input their API keys. This includes both cloud-based providers (e.g., OpenAI, Gemini, OpenRouter, Grog) and local models.
  2. Workspace: When selecting a provider or model in the workspace, users are presented with the entire list of available options, even if they have only configured a subset of these.

Issues:

  • Overwhelming Choices: Users have to sift through all available providers and models each time they want to make a selection, even if they only use a few.
  • Repetitive Selection: If a user wants to switch models, they must navigate through the entire list again, which is time-consuming and inefficient.
  • User Confidence: The abundance of options can be intimidating for regular users, who may fear making incorrect selections that could lead to unintended consequences, such as high costs or configuration errors.

Proposed Solution:

  • Streamlined Workspace Selection: In the workspace, only display the providers and models that the user has previously configured and successfully authenticated (i.e., those with working API keys).
  • Simplified Model Switching: Allow users to easily switch between their configured models without having to navigate through the full list of options.
  • User-Friendly Interface: Ensure that the interface is intuitive and reduces the cognitive load on users, making it easier for them to manage their AI providers and models without fear of making mistakes.

Benefits:

  • Improved User Experience: Users will find it easier and faster to select and switch between their preferred providers and models.
  • Reduced Complexity: By limiting the options to only those that are relevant to the user, the interface becomes less overwhelming, especially for non-technical users.
  • Enhanced Confidence: Users will feel more confident in managing their AI tools, knowing that they are only interacting with options that are already set up and working.

Example Scenario:

  • A user has configured API keys for OpenAI, Gemini, and OpenRouter, and has selected specific models from each provider.
  • In the workspace, the user should only see OpenAI, Gemini, and OpenRouter as provider options, and only the models they have selected from these providers.
  • If the user wants to switch from an OpenAI model to a Gemini model, they can do so quickly without having to navigate through unrelated options.

This revised explanation should make the issue clearer and highlight the need for a more user-friendly approach to provider and model selection in the desktop version.

@ozp commented on GitHub (Jan 29, 2025): ok, its more a UX issue I'm not good at english, so I asked deepseek for help: --- **Feature Request: Simplified Provider/Model Selection in Desktop Version** **Description:** The current process for selecting providers and models in the desktop version can be confusing and cumbersome, especially for regular users. While the full list of providers and models is necessary for initial setup and key configuration, the workspace experience should be streamlined to only show the providers and models that have been previously selected and configured by the user. **Current Workflow:** 1. **Settings:** Users can choose providers, models, and input their API keys. This includes both cloud-based providers (e.g., OpenAI, Gemini, OpenRouter, Grog) and local models. 2. **Workspace:** When selecting a provider or model in the workspace, users are presented with the entire list of available options, even if they have only configured a subset of these. **Issues:** - **Overwhelming Choices:** Users have to sift through all available providers and models each time they want to make a selection, even if they only use a few. - **Repetitive Selection:** If a user wants to switch models, they must navigate through the entire list again, which is time-consuming and inefficient. - **User Confidence:** The abundance of options can be intimidating for regular users, who may fear making incorrect selections that could lead to unintended consequences, such as high costs or configuration errors. **Proposed Solution:** - **Streamlined Workspace Selection:** In the workspace, only display the providers and models that the user has previously configured and successfully authenticated (i.e., those with working API keys). - **Simplified Model Switching:** Allow users to easily switch between their configured models without having to navigate through the full list of options. - **User-Friendly Interface:** Ensure that the interface is intuitive and reduces the cognitive load on users, making it easier for them to manage their AI providers and models without fear of making mistakes. **Benefits:** - **Improved User Experience:** Users will find it easier and faster to select and switch between their preferred providers and models. - **Reduced Complexity:** By limiting the options to only those that are relevant to the user, the interface becomes less overwhelming, especially for non-technical users. - **Enhanced Confidence:** Users will feel more confident in managing their AI tools, knowing that they are only interacting with options that are already set up and working. **Example Scenario:** - A user has configured API keys for OpenAI, Gemini, and OpenRouter, and has selected specific models from each provider. - In the workspace, the user should only see OpenAI, Gemini, and OpenRouter as provider options, and only the models they have selected from these providers. - If the user wants to switch from an OpenAI model to a Gemini model, they can do so quickly without having to navigate through unrelated options. --- This revised explanation should make the issue clearer and highlight the need for a more user-friendly approach to provider and model selection in the desktop version.
yindo commented 2026-02-22 18:27:20 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jan 30, 2025):

I understand. However, this will inadvertently lead to a worse UX if we do this. The reason is because let's say the user wants to use another provider totally. We should not make them:

  • Go to llm prefence page
  • Change provider just to input key into LLM Preference and save
  • Now swap the system back to what it was before so that is unchanged, but the provider is registered.
  • No you can go to workspace and use it.

In our current UX, we cover both the case of if you have never touched that provider and if you have used it since we can import the settings you may have already set.

I agree we can improve the model-switching within context but the reality is that if we constrain the user to already previously used providers we are force a very cumbersome UX to get it available.

We actually have designs and plans to move this around so it functions the way you desire, but its in a new form factor that will make more sense when that is live.

@timothycarambat commented on GitHub (Jan 30, 2025): I understand. However, this will inadvertently lead to a _worse UX_ if we do this. The reason is because let's say the user _wants_ to use another provider totally. We should not make them: - Go to llm prefence page - Change provider just to input key into LLM Preference and save - Now swap the system back to what it was before so that is unchanged, but the provider is registered. - No you can go to workspace and use it. In our current UX, we cover both the case of if you have never touched that provider **_and_** if you have used it since we can import the settings you may have _already_ set. I agree we can improve the model-switching within context but the reality is that if we constrain the user to already previously used providers we are force a very cumbersome UX to get it available. We actually have designs and plans to move this around so it functions the way you desire, but its in a new form factor that will make more sense when that is live.
yindo changed title from [FEAT]: easier method to choose provider / models (desktop) to [GH-ISSUE #3029] [FEAT]: easier method to choose provider / models (desktop) 2026-06-05 14:43:33 -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 enhancement feature request
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#1938
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?