danvics/haiku.rag
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Increase reranking candidate retrieval multiplier from 3x to 10x for improved result quality

Browse source
This commit is contained in:
Yiorgis Gozadinos 2025-11-18 14:35:57 +02:00
parent b1d19651d5
commit cc09ce87c2
No known key found for this signature in database
3 changed files with 5 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
CHANGELOG.md
View file

@ -1,6 +1,8 @@
# Changelog
## [Unreleased]
- Increase reranking candidate retrieval multiplier from 3x to 10x for improved result quality
## [0.17.0] - 2025-11-17
### Added

2
docs/configuration.md
View file

@ -476,7 +476,7 @@ See the [Pydantic AI documentation](https://ai.pydantic.dev/models/) for the com
## Reranking
Reranking improves search quality by re-ordering the initial search results using specialized models. When enabled, the system retrieves more candidates (3x the requested limit) and then reranks them to return the most relevant results.
Reranking improves search quality by re-ordering the initial search results using specialized models. When enabled, the system retrieves more candidates (10x the requested limit) and then reranks them to return the most relevant results.
Reranking is **disabled by default** (`provider: ""`) for faster searches. You can enable it by configuring one of the providers below.

4
haiku_rag_slim/haiku/rag/client.py
View file

@ -466,8 +466,8 @@ class HaikuRAG:
# No reranking - return direct search results
return await self.chunk_repository.search(query, limit, search_type, filter)
# Get more initial results (3X) for reranking
search_limit = limit * 3
# Get more initial results (10X) for reranking
search_limit = limit * 10
search_results = await self.chunk_repository.search(
query, search_limit, search_type, filter
)