Implementation:Infiniflow Ragflow DialogService Update Retrieval Settings

Knowledge Sources	RAGFlow
Domains	RAG, Information_Retrieval
Last Updated	2026-02-12 06:00 GMT

Overview

Concrete tool for configuring retrieval parameters on a chat application provided by RAGFlow DialogService.

Description

Uses DialogService.update_by_id to set retrieval fields on the Dialog model. These parameters are read at chat-time by Dealer.retrieval() to control hybrid search behavior.

Usage

Called via REST API to tune retrieval behavior for a chat application.

Code Reference

Source Location

• Repository: ragflow
• File: api/db/db_models.py (L950-955), api/db/services/dialog_service.py (L560-574)

Signature

DialogService.update_by_id(pid: str, data: dict) -> int
# data = {
#     "similarity_threshold": 0.2,
#     "vector_similarity_weight": 0.3,
#     "top_n": 6,
#     "top_k": 1024,
#     "rerank_id": "bge-reranker-v2-m3"
# }

Import

from api.db.services.dialog_service import DialogService

I/O Contract

Inputs

Name	Type	Required	Description
similarity_threshold	float	No	Minimum similarity (default 0.2)
vector_similarity_weight	float	No	Vector vs keyword weight (default 0.3)
top_n	int	No	Chunks for LLM context (default 6)
top_k	int	No	Initial retrieval pool (default 1024)
rerank_id	str	No	Reranking model ID

Outputs

Name	Type	Description
num	int	Rows updated

Usage Examples

from api.db.services.dialog_service import DialogService

DialogService.update_by_id("dialog-uuid-123", {
    "similarity_threshold": 0.3,
    "vector_similarity_weight": 0.5,
    "top_n": 10,
    "rerank_id": "bge-reranker-v2-m3"
})

Related Pages

Implements Principle

• Principle:Infiniflow_Ragflow_Retrieval_Configuration