Upload any product photo. Our AI detects objects with YOLO-World, encodes them with SigLIP 2, and fuses results with text search via Reciprocal Rank Fusion.
Open-vocabulary object localization identifies products in any photo — no pre-trained classes needed. Outputs bounding boxes for each detected item.
State-of-the-art vision encoder maps product images into a 1152-dim embedding space. Cross-modal alignment ensures text and image queries share the same vector space.
1152 dimensions · so400m-patch14-384
crop=true: YOLO detects → crops each object → embeds individually for precise per-item matching. crop=false: Embeds the full image for scene-level similarity.
Per-object embed
Scene-level embed
Text results (BGE-M3 dense + sparse) and image results (SigLIP 2 vision) are independently ranked, then merged via RRF to produce a single unified result set — capturing both semantic meaning and visual similarity.
V1 delivers fast keyword search. V2 adds a full NLP pipeline — NER, transliteration, spellcheck, synonym expansion, and reranking.
No results found for "xaxox"
Latin input not understood
Խաղող պրեմիում 1 կգ
Խաղող Պերու 1 կգ
Խաղող սեվ անկոշտ 1 կգ
Each layer works in concert to deliver results that feel like reading the shopper's mind.
SigLIP 2 vision embeddings + BGE-M3 text embeddings + YOLO object detection. Snap a photo and find it instantly via server-side RRF fusion.
Algorithmic "Arminglish" transliteration handles Latin-typed Armenian (xaxox → խաղող), slang dictionaries, recursive synonym expansion, and agglutinative stemming.
Dynamic brand and category boosts, trending scores, merchandising controls, real-time analytics, and "Did you mean?" spell-check to maximize conversion rates.
Every millisecond counts. Engineered for sub-50ms responses at scale.
Autocomplete Latency
Text Search (V1)
Product Embeddings
Uptime SLA
Take absolute control over your catalog’s visibility. Blend state-of-the-art AI semantics with your strategic business goals.
Boost at the Product, Brand, or Category level. Elevate high-margin collections or demote out-of-season brands instantly.
Inject a real-time trending_score (0.0 to 1.0) to dynamically push viral or fast-selling products to the top without breaking search relevance.
Our proprietary RRF seamlessly multiplies pure vector similarity with your custom business weights, ensuring relevant but profitable results.
All business rules and trending scores are cached in-memory, guaranteeing sub-millisecond ranking adjustments without hitting the database.
413× faster · Zero database round-trips
Every result is scored through a multiplicative pipeline that preserves AI relevance while amplifying your business priorities.
Access the full management dashboard to test search, manage synonyms, and monitor analytics.
Ready to supercharge your e-commerce search? For enterprise integrations and custom deployments, contact our core engineering team.
Contact CityIT Engineering