2
[Feature Request] Add detailed debug logging for RAG/LLM requests
Source: paperless-ngx/paperless-ngx#13644 · opened by @JP95Git
Description
The current debug logging for the RAG chat only logs the user's query, for example:
[DEBUG] [paperless_ai.chat] Document chat query: ...
This makes troubleshooting provider-side issues difficult. For example, when a RAG request takes an unusually long time, it is currently hard to determine whether the delay occurs inside Paperless or while waiting for the AI provider.
A similar feature request is here:
When debug logging is enabled, provide a detailed but safe log of AI provider requests and responses.
For example:
The logging should contain enough information to correlate requests with provider-side logs and diagnose latency or API errors, while avoiding sensitive data such as API keys, document contents, full prompts, or generated responses.
Other
_No response_
The current debug logging for the RAG chat only logs the user's query, for example:
[DEBUG] [paperless_ai.chat] Document chat query: ...
This makes troubleshooting provider-side issues difficult. For example, when a RAG request takes an unusually long time, it is currently hard to determine whether the delay occurs inside Paperless or while waiting for the AI provider.
A similar feature request is here:
When debug logging is enabled, provide a detailed but safe log of AI provider requests and responses.
For example:
The logging should contain enough information to correlate requests with provider-side logs and diagnose latency or API errors, while avoiding sensitive data such as API keys, document contents, full prompts, or generated responses.
Other
_No response_
No pledges yet. Be the first to back this.
Comments
Similar requests
[Feature Request] Optimize AI suggestion generation
5 votes · 0 comments
[Feature Request] Different keys for LLM vs. LLM embeddings
6 votes · 0 comments
[Feature Request] Logging for AI/LLM usage
4 votes · 0 comments
[Feature Request] Allow overriding the embeddings encoding_format for OpenAI-compatible LLM providers
4 votes · 0 comments
[Feature Request] AI chat: size the synthesis prompts by context window, not embedding chunk size
3 votes · 0 comments
No comments yet.