Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-20 14:43:37 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #4502] [FEAT]: [Desktop] Request for RAG performance diagnostics and detailed logging #2865

New Issue
Open
opened 2026-02-22 18:31:34 -05:00 by yindo · 0 comments
yindo commented 2026-02-22 18:31:34 -05:00
Owner
Copy Link
Copy Source

Originally created by @matteolavaggi on GitHub (Oct 7, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/4502

Context

I'm experiencing slow response times on the Desktop Edition of AnythingLLM and I'm trying to diagnose where the bottleneck is occurring. I suspect it's related to the RAG process rather than the LLM itself, but I lack visibility into the execution pipeline.

Current Situation

  • Setup: Desktop Edition with external LLM provider (to eliminate local model performance issues)
  • Observation:
    • Embedding process is very fast ⚡
    • Answer generation is significantly slower 🐌
    • Using short test messages to isolate RAG performance from LLM reasoning time

Problem

The current logs don't provide sufficient information to diagnose performance issues:

  • No timing information for different pipeline stages (embedding retrieval, context preparation, LLM call, etc.)
  • No visibility into which execution path is being used
  • Unable to determine if the slowdown is in RAG retrieval, vector search, or elsewhere

Questions

  1. Is there an advanced logging mode available in the Desktop Edition that I'm missing?
  2. What's the recommended way to profile/debug performance in the desktop app?
  3. Do I need to switch to the Docker image or build from source to access detailed performance metrics?
  4. Are there any hidden configuration options or environment variables for verbose logging?

Desired Feature

It would be extremely helpful to have:

  • Timing breakdowns for each stage of the RAG pipeline (retrieval, ranking, context assembly, LLM inference)
  • Verbose logging mode showing the execution path
  • Performance metrics accessible from the desktop app logs or UI

Environment

  • Version: Windows X86
  • OS: Win 11
  • LLM Provider: Openai nano 5
  • Edition: Desktop (installable app)
Originally created by @matteolavaggi on GitHub (Oct 7, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/4502 ### Context I'm experiencing slow response times on the **Desktop Edition** of AnythingLLM and I'm trying to diagnose where the bottleneck is occurring. I suspect it's related to the RAG process rather than the LLM itself, but I lack visibility into the execution pipeline. ### Current Situation - **Setup**: Desktop Edition with external LLM provider (to eliminate local model performance issues) - **Observation**: - Embedding process is very fast ⚡ - Answer generation is significantly slower 🐌 - Using short test messages to isolate RAG performance from LLM reasoning time ### Problem The current logs don't provide sufficient information to diagnose performance issues: - No timing information for different pipeline stages (embedding retrieval, context preparation, LLM call, etc.) - No visibility into which execution path is being used - Unable to determine if the slowdown is in RAG retrieval, vector search, or elsewhere ### Questions 1. **Is there an advanced logging mode** available in the Desktop Edition that I'm missing? 2. **What's the recommended way to profile/debug performance** in the desktop app? 3. Do I need to switch to the **Docker image** or **build from source** to access detailed performance metrics? 4. Are there any **hidden configuration options** or environment variables for verbose logging? ### Desired Feature It would be extremely helpful to have: - Timing breakdowns for each stage of the RAG pipeline (retrieval, ranking, context assembly, LLM inference) - Verbose logging mode showing the execution path - Performance metrics accessible from the desktop app logs or UI ### Environment - **Version**: Windows X86 - **OS**: Win 11 - **LLM Provider**: Openai nano 5 - **Edition**: Desktop (installable app)
yindo added the enhancementfeature request labels 2026-02-22 18:31:34 -05:00
yindo changed title from [FEAT]: [Desktop] Request for RAG performance diagnostics and detailed logging to [GH-ISSUE #4502] [FEAT]: [Desktop] Request for RAG performance diagnostics and detailed logging 2026-06-05 14:48:58 -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 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#2865
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?