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 #5333] [BUG]: Thinking block text going out of div #5025

New Issue
Closed
opened 2026-06-05 14:51:34 -04:00 by yindo · 1 comment
yindo commented 2026-06-05 14:51:34 -04:00
Owner
Copy Link
Copy Source

Originally created by @Sultan-papagani on GitHub (Apr 2, 2026).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5333

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Image

1 - text is going out of the div.

2 - also what is this font? guys please pick a normal font like gpt or gemini

3 - the "hotbar" or whatever its called is also useless. settings: great, discord: no, why would i want that as a permanent button, docs: no, give money button: also no. in order for someone to donate, it should be useful in the first place.

4 - the system prompt "Given the following conversation, relevant context, and a follow up que..." thing couldnt be removed. i delete it and save it, and when i come back its still there. im not even sure if its supposed to be a template (doesnt look like one) or if i actually cant delete it.

5 -Workspace system: its not great. i just want normal chat. there should be an option about it: workspace mode - normal mode.

6 - why not make the setup easier by giving us a "llama.cpp" option? you guys have "ollama," why not the actual product instead of the weird version? i know it is the same as "generic openai," but when i type llama, at least show the generic openai version or duplicate generic openai and label it as "llama.cpp."

7 -

Image

whats with this empty space here? it doesnt do anything except make the window smaller. or why does the sidebar have too much padding? actually, every panel is padded for no reason. it just doesnt feel like im chatting with an LLM. heck even the llama-server looks better than this!

8 - lack of navigation: on settings, the back button is the settings button itself, but on other windows its on the top left. why?

9 - irony: we were supposed to run local LLMs on it privately, but you guys made it as a chrome instance, so it wastes 350MB. i need that RAM bro unless you guys donate free DDR5.

10 - you cant click the thinking box to make it larger, you have to press the tiny ↓ arrow icon to make it bigger. why?

Are there known steps to reproduce?

No response

Originally created by @Sultan-papagani on GitHub (Apr 2, 2026). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5333 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? <img width="2559" height="1386" alt="Image" src="https://github.com/user-attachments/assets/84dac598-03ca-4a7a-b8cf-8192b08813f1" /> 1 - text is going out of the div. 2 - also what is this font? guys please pick a normal font like gpt or gemini 3 - the "hotbar" or whatever its called is also useless. settings: great, discord: no, why would i want that as a permanent button, docs: no, give money button: also no. in order for someone to donate, it should be useful in the first place. 4 - the system prompt "Given the following conversation, relevant context, and a follow up que..." thing couldnt be removed. i delete it and save it, and when i come back its still there. im not even sure if its supposed to be a template (doesnt look like one) or if i actually cant delete it. 5 -Workspace system: its not great. i just want normal chat. there should be an option about it: workspace mode - normal mode. 6 - why not make the setup easier by giving us a "llama.cpp" option? you guys have "ollama," why not the actual product instead of the weird version? i know it is the same as "generic openai," but when i type llama, at least show the generic openai version or duplicate generic openai and label it as "llama.cpp." 7 - <img width="2559" height="518" alt="Image" src="https://github.com/user-attachments/assets/46a4597b-b0f7-4bba-a931-4443f19e110f" /> whats with this empty space here? it doesnt do anything except make the window smaller. or why does the sidebar have too much padding? actually, every panel is padded for no reason. it just doesnt feel like im chatting with an LLM. heck even the llama-server looks better than this! 8 - lack of navigation: on settings, the back button is the settings button itself, but on other windows its on the top left. why? 9 - irony: we were supposed to run local LLMs on it privately, but you guys made it as a chrome instance, so it wastes 350MB. i need that RAM bro unless you guys donate free DDR5. 10 - you cant click the thinking box to make it larger, you have to press the tiny ↓ arrow icon to make it bigger. why? ### Are there known steps to reproduce? _No response_
yindo added the possible bug label 2026-06-05 14:51:34 -04:00
yindo closed this issue 2026-06-05 14:51:34 -04:00
yindo commented 2026-06-05 14:51:35 -04:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Apr 2, 2026):

1 - text is going out of the div.

Happy to look into this. Could you click "Edit" on the message and paste the exact LLM response? That'll help us reproduce it.

2 - also what is this font? guys please pick a normal font like gpt or gemini

Curious what looks off, could you share a screenshot or describe what you're seeing? The font is intentional but we're open to feedback.

3 - the "hotbar" or whatever its called is also useless. settings: great, discord: no, why would i want that as a permanent button, docs: no, give money button: also no.

AnythingLLM is a free project maintained by a small team. The hotbar reflects that, community support, documentation, and yes, the ability to contribute financially are what keep this project alive and actively developed. If none of that is useful to you, that's fine, but it's very useful to the people who make the software you're using and writing this about too.

4 - the system prompt "Given the following conversation..." thing couldnt be removed.

It's a template, not a fixed string, you can replace it with anything you like. We intentionally prevent it from being completely empty as having no system prompt tends to degrade response quality noticeably.

5 - Workspace system: its not great. i just want normal chat.

Workspaces are a core part of how AnythingLLM is designed, they let you scope context and organize content without re-explaining things each session. A lot of users find this valuable. A "simple mode" is something we can consider if there's enough demand.

6 - why not make the setup easier by giving us a "llama.cpp" option?

Ollama and llama.cpp have diverged quite a bit and aren't interchangeable anymore. The Generic OpenAI provider is the catch-all for any OpenAI-spec endpoint including llama.cpp.

whats with this empty space here? every panel is padded for no reason.

The spacing is a deliberate design choice to keep the UI clean and readable. If you have specific mockups or suggestions we're genuinely open to seeing them.

8 - lack of navigation: on settings, the back button is the settings button itself, but on other windows its on the top left.

Settings entry and exit are intentionally in the same place. Could you clarify which page you're on when the back button appears top-left? That would help us identify if there's a real inconsistency worth fixing.

9 - you guys made it as a chrome instance, so it wastes 350MB.

Electron uses Chromium, not Chrome - different things. The tradeoff is a genuinely cross-platform experience without maintaining separate native apps. Given that a capable local LLM itself runs 3–5GB, the overhead is small in context.

10 - you cant click the thinking box to make it larger, you have to press the tiny ↓ arrow icon.

Clicking the thinking box should expand it; is that not working for you? If clicking isn't registering that sounds like a bug worth investigating.

<!-- gh-comment-id:4180500968 --> @timothycarambat commented on GitHub (Apr 2, 2026): > 1 - text is going out of the div. Happy to look into this. Could you click "Edit" on the message and paste the exact LLM response? That'll help us reproduce it. > 2 - also what is this font? guys please pick a normal font like gpt or gemini Curious what looks off, could you share a screenshot or describe what you're seeing? The font is intentional but we're open to feedback. > 3 - the "hotbar" or whatever its called is also useless. settings: great, discord: no, why would i want that as a permanent button, docs: no, give money button: also no. AnythingLLM is a free project maintained by a small team. The hotbar reflects that, community support, documentation, and yes, the ability to contribute financially are what keep this project alive and **actively** developed. If none of that is useful to you, that's fine, but it's very useful to the people who make the software you're using and writing this about too. > 4 - the system prompt "Given the following conversation..." thing couldnt be removed. It's a template, not a fixed string, you can replace it with anything you like. We intentionally prevent it from being completely empty as having no system prompt tends to degrade response quality noticeably. > 5 - Workspace system: its not great. i just want normal chat. Workspaces are a core part of how AnythingLLM is designed, they let you scope context and organize content without re-explaining things each session. A lot of users find this valuable. A "simple mode" is something we can consider if there's enough demand. > 6 - why not make the setup easier by giving us a "llama.cpp" option? Ollama and llama.cpp have diverged quite a bit and aren't interchangeable anymore. The Generic OpenAI provider is the catch-all for any OpenAI-spec endpoint including llama.cpp. > whats with this empty space here? every panel is padded for no reason. The spacing is a deliberate design choice to keep the UI clean and readable. If you have specific mockups or suggestions we're genuinely open to seeing them. > 8 - lack of navigation: on settings, the back button is the settings button itself, but on other windows its on the top left. Settings entry and exit are intentionally in the same place. Could you clarify which page you're on when the back button appears top-left? That would help us identify if there's a real inconsistency worth fixing. > 9 - you guys made it as a chrome instance, so it wastes 350MB. Electron uses Chromium, not Chrome - different things. The tradeoff is a genuinely cross-platform experience without maintaining separate native apps. Given that a capable local LLM itself runs 3–5GB, the overhead is small in context. > 10 - you cant click the thinking box to make it larger, you have to press the tiny ↓ arrow icon. Clicking the thinking box should expand it; is that not working for you? If clicking isn't registering that sounds like a bug worth investigating.
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#5025
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?