Config reference: sparse_search

• Enterprise tuning surface

  ---

  Defaults + constraints are rendered directly from Pydantic.

• Env keys when available

  ---

  Many fields have an env-style alias (from TriBridConfig.to_flat_dict()).

• Tooltip-level guidance

  ---

  If a matching glossary entry exists, you’ll see deeper tuning notes.

Config reference Config API & workflow Glossary

Total parameters: 4

Group index

• (root)

(root)

JSON key	Env key(s)	Type	Default	Constraints	Summary
sparse_search.bm25_b	—	float	0.4	≥ 0.0, ≤ 1.0	BM25 length normalization for the Qdrant/bm25 sparse vectors (0 = no penalty, 1 = full penalty). Part of the sparse index contract (re-index on change).
sparse_search.bm25_k1	—	float	1.2	≥ 0.5, ≤ 3.0	BM25 term-frequency saturation for the Qdrant/bm25 sparse vectors (higher = more weight to term frequency). Part of the sparse index contract (re-index on change).
sparse_search.enabled	—	bool	true	—	Enable the sparse (Qdrant/bm25) leg in tri-brid retrieval
sparse_search.top_k	—	int	50	≥ 10, ≤ 200	Number of results to retrieve from the sparse leg