krzysztof/landing_page
1
0
Fork 0
mirror of https://github.com/qdrant/landing_page.git synced 2026-09-30 00:18:32 +02:00
Code Issues Packages Projects Releases Wiki Activity

Fix ANN precision tutorial: tune hnsw_ef, not build-time params

Browse Source 
The Search Quality tab uses the Query Points API, so it can only
tune search-time SearchParams. Replace the m/ef_construct walkthrough
with hnsw_ef and link to the Essentials course for the build-time
trade-offs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Dylan Couzon
2026-04-29 14:49:05 -04:00
co-authored by Claude Opus 4.7
parent ac9f77d9f3
commit b7b2865e14
1 changed files with 6 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files
qdrant-landing/content/documentation/tutorials-search-engineering/retrieval-quality.md
+6 -11
View File
@@ -11,8 +11,7 @@ weight: 5
| Time: 15 min | Level: Beginner | | |
|--------------|---------------------|--|----|
This tutorial focuses on **ANN precision**: how closely approximate nearest-neighbor (ANN) search matches exact kNN search.
To measure ANN precision, you compare Qdrant's approximate top-k against the exact kNN top-k using `precision@k`, then tune HNSW parameters to trade memory and build time for higher precision.
This tutorial focuses on **ANN precision**: how closely approximate nearest-neighbor (ANN) search matches exact kNN search, measured with `precision@k`.
**Prerequisites.** A Qdrant collection populated with your documents as points (vectors + optional payload).
@@ -33,23 +32,19 @@ Qdrant's Web UI includes a Search Quality tab that measures the gap between appr
![Search Quality tab with default evaluation results](/documentation/tutorials/retrieval-quality/search-quality-tab.png)
The tab reports average **precision@k** (1.0 = perfect overlap; 0.95+ is typical for well-tuned HNSW). HNSW has tunable parameters that trade memory and index build time for higher precision.
The tab reports average **precision@k** (1.0 = perfect overlap; 0.95+ is typical for well-tuned HNSW).
## Tuning the HNSW Parameters
## Tuning Search Precision
HNSW is a hierarchical graph where each node has a set of links to other nodes. The `m` parameter controls the number of edges per node: higher `m` means higher precision at the cost of more memory. The `ef_construct` parameter controls how many neighbors are considered during index building: higher `ef_construct` means higher precision at the cost of longer indexing time. The defaults are `m=16` and `ef_construct=100`.
For the full list of HNSW parameters, including on-disk storage and precision/memory trade-offs, see [Optimize Performance](/documentation/ops-optimization/optimize/).
Toggle **advanced mode** in the Search Quality tab to tune these parameters inline. To see the effect, try raising `m` and `ef_construct` (for example, to 32 and 200), then run the evaluation again.
Toggle **advanced mode** in the Search Quality tab to tune search-time parameters inline. The main one is `hnsw_ef`: the number of candidates evaluated during a search. Raising it explores more of the graph, improving precision at the cost of higher query latency. To see the effect, raise `hnsw_ef` (for example, to 256) and run the evaluation again.
![Search Quality advanced mode with HNSW parameters](/documentation/tutorials/retrieval-quality/search-quality-advanced.png)
Precision should increase at the cost of higher build time and memory.
Precision should increase at the cost of higher query latency.
![Search Quality results after HNSW tuning](/documentation/tutorials/retrieval-quality/search-quality-after-tuning.png)
Tune until the balance between precision and cost matches your targets.
If `hnsw_ef` alone does not get you to your precision target, the build-time parameters `m` and `ef_construct` set the ceiling on how precise approximate search can be. Changing them requires rebuilding the HNSW index. For the trade-offs and how to choose values, see [HNSW Indexing Fundamentals](/course/essentials/day-2/what-is-hnsw/) in the Qdrant Essentials course.
## Automate in CI with Python